Table of Contents

Title page
KEY TO SYMBOLS
INTRODUCTION

CHAPTER 1. CLASSICAL VARIATION (Be2)

Part 1 Part 2 Part 3 Part 4

CHAPTER 2. 7.Bc4: ANTI-YUGOSLAV VARIATION

Part 1 Part 2 Part 3 Part 4 Part 5

CHAPTER 3. 7.Bc4: MY SYSTEM

Part 1 Part 2 Part 3

CHAPTER 4. MAROCZY BIND: BREYER VARIATION

Part 1 Part 2 Part 3 Part 4 Part 5 Part 6 Part 7 Part 8

CHAPTER 5. MAROCZY BIND: MAIN LINE

Part 1 Part 2 Part 3 Part 4 Part 5 Part 6

CHAPTER 6. 4.Qxd4 VARIATION

Part 1 Part 2

CHAPTER 7. ANTI-SICILIANS: ALAPIN AND MORRA

Part 1 Part 2 Part 3 Part 4 Part 5 Part 6

AFTERWORD

1
 THE HYPER ACCELERATED website: www.thinkerspublishing.com
DRAGON

by
Raja Panjwani

www.thinkerspublishing.com

Managing Editor
Romain Edouard

Proofreading
Daniël Vanheirzeele

Graphic Artist
Philippe Tonnard

Cover design
Iwan Kerkhof

Typesetting
i-Press ‹www.i-press.pl›

Second extended edition 2018 by Thinkers

Publishing

The Hyper Accelerated Dragon

Copyright © 2018 Raja Panjwani

All rights reserved. No part of this

publication may be reproduced, stored in a
retrieval system or transmitted in any form or by
any means, electronic, mechanical, photocopying,
recording or otherwise, without the prior written
permission from the publisher.

ISBN 978-94-9251-034-1
D/2018/13730/16

All sales or enquiries should be directed to

Thinkers Publishing, 9850 Landegem, Belgium.

e-mail: info@thinkerspublishing.com
2
Key to Symbols used!

! a good move
? a weak move
!! an excellent move
?? a blunder
!? an interesting move
?! a dubious move
□ only move
= equality
∞ unclear position
⩲ White stands slightly better
⩱ Black stands slightly better
± White has a serious advantage
∓ Black has a serious advantage
+- White has a decisive advantage
-+ Black has a decisive advantage
→ with an attack↑with an initiative
⇆ with counterplay
Δ with the idea of
⌓ better is
≤ worse is
N novelty
+ check
# mate
© with compensation for thesacrificed material

3
 INTRODUCTION
It simply isn’t an adventure worth telling if
there aren’t any dragons.
J.R.R. Tolkien
My Favorite Sicilian
I was introduced to the Accelerated Dragon
when I was ten years old, more than seventeen
years ago. It was arguably my first ‘serious’
defense against 1.e4: prior to then I would develop
my pieces in a manner my father and I called
‘P-Play’ (the ‘P’ deriving from our family name)
but which I later discovered is widely known as
the Hippopotamus Defence. He and I were of
similar strength at the time, and we studied the
opening together from the then recently published,
and now classic, Accelerated Dragons by IMs
Donaldson and Silman.
What drew me to the opening initially was
the abundance of cheapos I could set up for my
opponents in the early stages of the game, which
even experts and masters seemed unprepared for.
The following was always one of my favorites:
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6 6.Be3 Nf6 7.Bc4 Qa5 8.Qd2?
8.0-0
8...Nxe4! 9.Nxc6
9.Nxe4 Qxd2+ 10.Bxd2 Nxd4µ
4
9...Qxc3!! 10.Qxc3
10.bxc3 Nxd2 11.Bxd2 bxc6µ
10...Bxc3+ 11.bxc3 bxc6µ
Beyond simple tricks like this one, I found
that the positional themes of the opening were
fairly easy to digest; for example, dark square
control, central breakthrough (especially ...d7-d5),
queenside expansion, as well as the typical
favorable and unfavorable endgames which tend to
arise. As I have matured as a player, my
perspective on this opening has correspondingly
transformed, but my respect and appreciation for
its strength has only been enhanced. This book is
an attempt to convey my current understanding
and approach with black.
I have always felt that the Accelerated
Dragon does not get its due respect among the
Sicilians. Even its prodigal brother, the
un-accelerated Dragon, had its time in the
spotlight when it was used by Kasparov to twice
defeat (and twice draw) Anand in their 1995 PCA
World Championship match. Why then, has the
Accelerated Dragon — the theoretically no worse
off, and much safer of the two (I like to think of it
as the only Sicilian where Black needn’t worry
about getting mated in 25 moves) — historically
been only an occasional guest in top events, and,
unlike every other respectable Sicilian, never
occupied the central battlefield of a World
Championship match?1 Part of the discrepancy is
a vestige of the old (pre-1970s) dogma that in the
Sicilian, to avoid suffocation, Black must prevent
White from obtaining a ‘clamp’ pawn center
(pawns on e4 and c4). Indeed, the Maroczy Bind
(5.c4) has always been the bane of the Accelerated
Dragon’s existence. However, while this attitude
towards the Sicilian may have been justifiable half
a century ago, Black has since demonstrated
counterattacking prospects against the e4/c4 clamp
in a variety of structures, as in the Hedgehog,
Kalashnikov, Kan, Taimanov, and certainly no less
in the Accelerated Dragon.
I suspect that computers have deterred
many potential devotees away from the
Accelerated Dragon. Computer evaluations in the
main lines tend to fluctuate between +0.25 and
+0.5, which plausibly leads to the rationale that
playing the Accelerated Dragon instead of the
Berlin or Marshall — where evaluations are closer
to +0.15 — is like playing with a small handicap
straight out of the opening. Things, however, are
not so simple. Computers evaluate each position
by objective features, without regard for subjective
factors which are very often more important in
tournament chess. Machines systematically ignore
the value of, for example, being able to follow one
of a small number of thematic plans, irrespective
of what the opponent does, saving on clock time as
well as risk of mishandling the position. This sort
of human element is unaccounted for by the
engine, resulting in an inflated estimation of
White’s chances. In this regard, there are
similarities between the Accelerated Dragon and
the King’s Indian Defense — another opening
notoriously bastardized by the engine. King’s
Indian devotees are used to seeing +0.5 computer
evaluations, but they are not discouraged because
they recognize that there is a narrow margin of
error for White, and to err is human. The same can
be said for the Accelerated Dragon.
Fortunately, the tide of fashion is turning,
and contemporary Accelerated Dragon experts like
(super) Grandmasters Tiviakov, Mamedov,
Iturrizaga, and Malakhov have demonstrated that
this opening can be a reliable counter to 1.e4 even
against top opposition. Recently, in fact, World
Champion Magnus Carlsen upheld the Black side
of a Maroczy to put a halt to Caruana’s 7-0 run in
the 2014 Sinquefield Cup. I predict a bright future
for this opening, for many reasons, but most of all
because 1) The resulting positions are difficult for
computers to properly assess- many ‘+=’
5
evaluations are more accurately ‘=’ but more
importantly ‘easier to play for Black’, a factor
which should not be underestimated especially
considering the increasingly short time controls,
and 2) There is plenty of unexplored terrain, which
cannot be easily navigated by the positional
dictums we are brought up with, because the
Accelerated Dragon is a genuinely nonstandard
opening. This means that there is a competitive
advantage to those who work out its unusual
nuances, unlike in, say, the Najdorf or Sveshnikov
where it often feels like the strategic ideas are all
well known, and only concrete novelties are yet to
be discovered (if it is unclear what I mean by this,
I hope it isn’t by the end of the book!).
The Accelerated Dragon State of Mind
A friend of mine (a strong IM) recently
commented to me that if he could be certain that
his opponents wouldn’t play the Maroczy bind, he
would always play the Accelerated Dragon instead
of his usual (un-accelerated, but I sometimes
teasingly prefer ‘un-playable’) Dragon, because
White can’t play the critical Yugoslav Attack
against the Accelerated Dragon (despite this being
lesson #1 of the Accelerated Dragon, a surprising
number of masters have not gotten the memo).
“However”, he continued, “in the Maroczy, Black
is just playing for a draw, you can never win!” A
loyal defender of my beloved pet opening, I
insisted he had it all wrong, and that I welcome the
Maroczy in must-win games with Black. “That’s
really weird dude, you’re probably the only one”
was his retort, but I think when it comes to the
Accelerated Dragon, there’s a requisite state of
mind needed in order to properly handle it —
some players have had a conversion experience
after catching a glimpse of its incredible power,
while others haven’t. Plausibly as a result of this,
from my experience there is a peculiar
camaraderie among Accelerated Dragon
practitioners. Whereas Najdorf ‘bros’ espouse a
Darwinian angst that their novelty on move 25 in
the Poisoned Pawn variation will be discovered,
used, and rendered useless by their colleagues, I
have found that Accelerated Dragon players enjoy
discussing their ideas with each other. A personal
anecdote of mine is fairly typical: in the final
round of the 2013 US Masters tournament I was in
a must-win ‘money game’ with Black against
Cuban GM Abreu, and I noticed GM Rauf
Mamedov (a leading expert on the Black side of
the Accelerated Dragon) was taking an interest in
the Maroczy Bind on my board. I won the game in
a tense struggle, and afterwards when I was
collecting my prize, Rauf kindly congratulated me
on the win and took an interest in the 15...e6 line I
played (see chapter 5), which he said he hadn’t
studied before. I told him I was not too happy with
the more popular 15...Qb6, but he asserted that
from his analysis Black has no problems there —
“it’s equal” he said. His confident proclamation
was just the nudge I needed to look closer into
some of the lines I thought were undesirable for
Black, and on closer inspection I realized
(unsurprisingly) he was right!
While this elusive ‘state of mind’ is
somewhat ineffable, and better grasped from
experience than anything else (if I am successful
then the contents of this book will convey
precisely this), I think it is helpful to think of the
Maroczy as a close cousin of the Hedgehog. I
understand the ‘philosophy’ of the Hedgehog in
terms of how Mihai Suba describes it in his
excellent Dynamic Chess Strategy. It is worth
quoting him at length.
“White’s position looks ideal. That’s the
naked truth about it, but the ‘ideal’ has by
definition one drawback — it cannot be improved.
...In the early 1970s, the successes of Karpov and
Andersson showed that [Hedgehog] positions are
not only playable but offer as many winning
chances as any other opening. This was in glaring
conflict with classical strategy. White enjoys more
space, better development [and] his position has
no weaknesses. How is it possible that Black not
only resists in these positions but sometimes wins?
The only plausible answer lies in the hidden
dynamics of the positions. After the opening,
White’s position has all the qualities of a
successful picture, but lacks concrete possibilities
for improvement. Within our terminology, it is
rigid (not elastic). Black’s position, in contrast,
‘looks’ bad but has greater scope for
improvement.” (p. 26)
This description applies equally well to the
Maroczy. In fact, you might say that the
6
Accelerated Dragon (specifically the Black side of
the Maroczy) ‘state of mind’ is, to borrow Suba’s
phrase, an appreciation for the “hidden dynamic
factors” in each position which compensate for the
static deficiencies (again, the best way to ‘sense’
these is by studying the opening — the variations
in this book are meant to illustrate these factors).
Moreover, as a long time Hedgehog player myself,
I must say that in my opinion, Black has much
more freedom in the Maroczy than in the
Hedgehog; for instance, in the Hedgehog, it is
usually unfavorable for Black to exchange queens,
whereas in the Maroczy (and the Accelerated
Dragon more broadly), White often takes pains to
avoid exchanging queens so as to not lose the
initiative, and that is a liability which contributes
to the “rigidity” (another apt term of Suba’s) of
White’s position. Terms like ‘elastic’ and
‘counterattacking potential’ will be interspersed
throughout this book — they are much more
informative than reductive evaluations like ‘=’.
An Inclusive Opening
One of the remarkable things about the
Accelerated Dragon is its appeal to players with
vastly different styles. Compare Bent Larsen, the
epitome of dynamic, offbeat, risky chess, with
Sergei Tiviakov, who claimed in an interview
recently that his style has been shaped most by
Petrosian (who was a great Accelerated Dragon
devotee himself), Smyslov, and Karpov — both
these players have championed the Accelerated
Dragon as their main weapon against 1.e4 and yet
their styles are in many ways polar opposites of
each other! How can this be? I think the answer to
this question is subtle and instructive. I think that
when playing the Accelerated Dragon it ‘feels’
like you are playing White, not Black (albeit in a
hypermodern manner). What I mean is, in chess,
White tends to be the one to control the tempo of
the game — usually it is White who chooses
whether to enter into an opposite side castling
situation, or to exchange pieces early on and
maneuver around in a simplified middlegame, or
invoke the center as the locus of battle, ensuring
king safety above all.2 Furthermore, Black usually
needs to play accurately to not end up slightly
worse, or at least give the initiative to White. The
situation is, to the well prepared Accelerated
Dragon player, precisely the reverse: in the
Maroczy, for example, there are half a dozen
different ways for Black to develop, and players of
diverse styles can choose the one which suits them
best (or vary their choice depending on practical
considerations). Black controls the tempo and
determines the character of the struggle, which is
why it is so effective in must-win games.
Furthermore, unlike in many 1...e5 openings, or in
most other Sicilians like the Kan, Sveshnikov, or
even the Najdorf, White’s choices are rather
limited if he does not want to end up slightly
worse out of the opening. In practice, White meets
the Accelerated Dragon with either the Maroczy
Bind or the 7.Bc4 variation; this is simply not so in
the Najdorf where every single reasonable move is
a viable candidate from the starting position of the
Najdorf (6.h3, 6.Rg1, 6.g3, 6.f3, 6.Be3, 6.f4,
6.Be2, 6.Bc4, 6.Bg5, 6.a4, and that is not even to
mention variations therein), and the margin for
White error is far greater (for example 6.Be2 e5
7.Nf3!? is a serious challenge to the Najdorf but
6.Be2 Bg7 7.Nf3?! is just dubious against the
Accelerated Dragon).
The above may sound a little hyperbolic,
and I would like to make it clear from the outset
that I am not claiming that ‘Black is better’ in the
Accelerated Dragon; to do so would be dishonest.
My claim is a serious one: the character of the
Accelerated Dragon is that of a White opening. In
fact, the Accelerated Dragon reversed is a White
opening, called the English, and is fashioned by
most of the top players in the world, including
Carlsen, Kramnik, Aronian, Anand, Giri, and
others: the exact piece arrangement occurs with
colors reversed (and a tempo up) after 1.c4 e5 2.g3
Nf6 3.Bg2 d5 4.cxd5 Nxd5 5.Nc3,
as well as 1.c4 c5 2.Nf3 Nf6 3.Nc3 d5
4.cxd5 Nxd5 5.g3 Nc6 6.Bg2 Nc7 7.0-0 e5
(reversed Maroczy). I have enjoyed playing this
‘reversed Accelerated Dragon’ with White as well.
In addition to being inclusive in the above
sense, that it can suit players of diverse styles, and
also that it can be a coherent complement to a 1.c4
repertoire with White, there are many interesting
ways that the opening ‘fits’ with defenses against
1.d4/1.c4/1.Nf3, and can often directly transpose
from them.
7
King’s Indian:
1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6
i) 5.Nf3 0-0 6.Be2 c5 7.0-0 cxd4 8.Nxd4
Nc6
ii) 5.f3 0-0 6.Be3 c5 7.Nge2 cxd4 8.Nxd4
Nc6
Benoni/Benko Gambit:
1.d4 Nf6 2.c4 c5 3.Nf3 cxd4 4.Nxd4 Nc6
5.Nc3 g6 6.e4
1.d4 Nf6 2.Nf3 g6 3.c4 c5 4.Nc3 (4.d5
either 4...b5 or 4...e6) 4...cxd4 5.Nxd4 Nc6 6.e4
Symmetrical English:
1.c4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.e4 Nc6
1.Nf3 c5 2.c4 g6 3.d4 cxd4 4.Nxd4 Bg7
5.e4 Nc6
Of course, there is no obligation on
Accelerated Dragon players to deploy these
defences in order to allow for transpositional
possibilities — Tiviakov has been a lifelong
Nimzo-Indian/Queen’s Indian devotee as a
counterexample — but I have found it useful to
play these systems in tandem myself.
One more point on the topic of move
orders: since the Accelerated Dragon (especially
the Maroczy) can come about from so many
different move orders, I have taken some liberties
with the games in this book to convert the initial
moves to the 1.e4 c5 2.Nf3 g6 ‘Hyper-Accelerated
Dragon’ move order we will be focusing on. I
have done this, following a not uncommon
practice, purely for didactic purposes — I don’t
want readers happy with their 1.d4 defenses to be
confused by transpositions from openings they
don’t play.
About This Book
This book presents a repertoire for Black
after 1.e4 c5 2.Nf3 g6 — the Hyper-Accelerated
Dragon. I am relatively lax about distinguishing
between ‘Accelerated Dragon’ (1.e4 c5 2.Nf3 Nc6
3.d4 cxd4 4.Nxd4 g6) and ‘Hyper-Accelerated
Dragon’, and I use the two interchangeably unless
to emphasize move order nuances, for example,
“the Hyper-Accelerated Dragon avoids the
Rossolimo”. However, this is not merely ‘a’
Hyper-Accelerated Dragon repertoire, it is my
repertoire, and I present the material as such, from
a first person perspective, making brazen use of
my own games and offering personal anecdotes
and opinions. This stylistic choice risks my
coming across as presumptuous and at times even
boastful, but my hope is rather that the
conversational mode of presentation makes readers
feel as though I am their tour guide through what
might otherwise feel like an insurmountable
labyrinth of variations. Further on the point of
stylistic choices, I am regrettably not sufficiently
skilled in writing without gender-specific
pronouns, so please regard all generic references
to ‘he’ as ‘s/he’ (or alternative) and so on.
I provide as much information as I think is
necessary for readers to play this opening with
Black; however, this is absolutely not meant to be
an anthology on the opening. I make no claim to
cover every conceivable variation White can play.
Any attempt at such, couched under the heading of
a ‘complete repertoire’ would not only be
misleading, but in this day and age obsolete. This
is not to say that opening books are altogether
obsolete; on the contrary, as inundated with
information as we all are nowadays, it can be
enormously helpful to have an author divulge
opening secrets from their years of experience
which would not easily be gathered from a
database search.
What is obsolete is the attempt to
thoroughly and comprehensively ‘prove equality’
with Black, and more importantly for our purposes
it is antagonistic to the spirit of the Accelerated
Dragon, which is that of an opening refusing to be
evaluated on static grounds alone. As Jonathan
Rowson instructs in his Seven Deadly Chess Sins,
“You need to assess not only the position as it
stands but the position as it has changed and how
it is likely to continue to change”. (p.75) So, I am
not a big fan of evaluations like ‘=’ or ‘=+’ or ‘+=’
(though I capitulate to these at times) because ‘=’
makes me think of a draw and ‘+=’ makes me feel
like I ought to be satisfied with a draw as Black,
when in reality Black can very much be optimistic
about his position despite such evaluations, and
that is why I prefer evaluations like “counterplay”
or “mutual chances”.
If you encounter a variation not covered in
this book, for example 1.e4 c5 2.Na3, my general
prescription is this: find a database (no excuses,
8
they are free online), and search the position with
an Accelerated Dragon player (I gave you a list
above) as Black to see how they have chosen to
play the position. Ideally you’ll find a model game
that you can recall whenever you face the system;
after all, when it comes to rare systems like 2.Na3
it is foolish to memorize concrete variations since
you’ll never remember them anyway, but the key
ideas of a model game you can. Even in the main
lines of the Accelerated Dragon, don’t try to
memorize the moves given in this book as if they
are the ultimate truth. They aren’t. Your learning
will be enhanced if you actively seek out novelties
of your own, and try to understand how the
various positional ideas for both sides fit together.
To make this book as valuable to the
Petrosian-style (risk-averse) Accelerated Dragon
player as to the Larsen-type (risk-craving), I have
recommended two systems against each of the
7.Bc4 and Maroczy Bind variations. I hope at least
one of these suits you. For those among you who
will embark on the risky course (‘My System’
against 7.Bc4 and the Breyer Variation of the
Maroczy), may I caution you to do so with a
realistic attitude towards the cost of risk-taking in
chess. It is in the (mathematical) nature of
risk-taking that it increases the variability of
outcomes — both good and bad. The mature risk
taker is mindful of this, cognizant that their
risk-taking is compatible with their aims and
justifications. This was the attitude of Bent Larsen.
If you would like to play the Accelerated Dragon
ambitiously, with a tolerance for risk, keep in
mind the following description of Larsen, given by
Reshevsky: “He is a firm believer in the value of
surprise. Consequently, he often resorts to dubious
variations in various openings. He also likes to
complicate positions even though it may involve
considerable risk. He has a great deal of
confidence in his game and fears no one. His
unique style has proven extremely effective
against relatively weak opponents but has not been
too successful against top-notchers.” Alas, this is
the risk-taker’s predicament, but far from
discouraging it, I am thankful for the risk-takers
among you who resist the ‘genetic’ drift of our
chess community towards timidity and
results-oriented pragmatism.
I sincerely hope you find this to be an
enjoyable and enriching experience.

9
 CHAPTER 1
CLASSICAL VARIATION (Be2)
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6
Contents
1. 6.Nb3, 6.Nde2
2. 6.Be3 Nf6 7.sidelines & 7.Be2 d5!?
3. 6.Be3 Nf6 7.Be2 0-0 8.sidelines
4. 6.Be3 Nf6 7.Be2 0-0 8.0-0
1
We begin with the variation which
epitomizes the Accelerated Dragon philosophy. In
most defenses to 1.e4, White has the option of
playing a ‘Classical’ variation by developing the
light squared bishop to e2. This is particularly the
case in Sicilians like the Najdorf, Scheviningen
and Taimanov, but also in other defenses like the
Pirc/Modern and Alekhine. In all these systems,
the Classical Variation offers White serious
chances to obtain an opening advantage, and the
10
immortal games of past champions like Geller and
Karpov provide textbook illustrations of successful
‘Classical’ 1.e4 play. In contrast, the Be2
(Classical) variation against the ordinary,
un-Accelerated Dragon is rather harmless for
Black, the Yugoslav Attack being its critical test.
As Accelerated Dragon players, we are in an even
more favorable situation than ordinary Dragon
players when it comes to the Classical variation,
because we can choose to transpose to harmless
variations of the Classical Dragon by opting for
...d7-d6 at opportune moments, and in some lines
we can strike with ...d7-d5 directly, saving a full
tempo compared with analogous lines in the
Dragon. Despite the fact that this system offers
White no advantage, it is still seen in about 15% of
Accelerated Dragons (the other 85% are nearly
evenly distributed between the Maroczy and Bc4
variations), the bulk of which occur at the club
level.
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6
6.Nb3
This move tends to be played later on
anyway in the Be2 system, in order to hinder
Black from playing ...d7-d5 (note that the Maroczy
and Bc4 variation both target the d5-square), so
some players prefer to play 6.Nb3 directly without
committing the Bishop to e3.
6.Nde2
 This is not part of the Be2 system so I just
mention it in passing. Some people who like to
fianchetto their light bishop in other Sicilians play
this ‘Chameleon’ line; such variations are apt
against tamer Sicilians like the Najdorf, not the
unforgiving Accelerated Dragon.
6...Nf6 7.g3
(7.a4 d5!N 8.exd5 (8.Nxd5? Nxe4µ)
8...Nb4 9.Nf4 (9.Ng3 0-0 10.Bc4 Qc7 11.Bb3
Rd8„) 9...Bf5
10.Bb5+ Kf8 11.Bd3 Nxd3+ 12.cxd3 g5
13.Nfe2 Nxd5 14.Bxg5 Nxc3 15.bxc3 Bxd3
16.0-0 Bc4=)
7...b5! 8.Bg2 Rb8 9.0-0 0-0
11
10.h3
a) 10.Nd5 d6 11.Bg5 (11.h3 Nd7 12.c3 e6
13.Nb4 Nxb4 14.cxb4 Nb6³ Polgar,Z
(2550)-Georgiev,V (2615) Matinhos 1994)
11...Nd7 12.c3 Re8 13.Nd4 Nxd4 14.cxd4 Nb6
15.Nb4 Bb7 16.Rc1 Qd7 17.b3 a5 18.Nc2 b4= 1–0
(57) Kuzmin,G (2540)-Macieja,B (2460) St
Petersburg 1996;
b) 10.Bf4 d6 11.Qd2 b4 12.Nd5 Ng4
13.Rab1 Nge5 14.b3 e6 15.Ne3 Qa5µ;
10...b4 11.Nd5 Ba6 12.Re1 Nxd5 13.exd5
Na5
14.Bf4
(14.Nd4 Nc4 15.Rb1 e5! 16.dxe6 fxe6³)
14...Rc8 15.d6 e6 16.a3 b3! 17.cxb3 Qb6
18.b4 Nc4³ Perovic-Nikolic, Pula 1991.
6...Nf6 7.Be2 0-0 8.0-0
There are ways of playing this in the spirit
of the Accelerated Dragon but I think Black’s best
and simplest path is to proceed in Dragon style
with ...d7-d6 and ...Bc8-e6, aiming to play
...d6-d5.
8...d6
9.Bg5
This is likely to be White’s idea if they
play 6.Nb3 directly.
9.f4?!
9...b5!„;
9.Kh1 Be6 10.f4 Qc8! Preventing f4-f5.
(10...d5?! 11.f5 Bc8 12.exd5 Nb4 13.fxg6
hxg6 14.Bf3 Bf5 15.Nd4±)
11.Be3
(11.f5? gxf5 12.exf5 Bxf5µ One might
think White has some compensation because
Black’s king position has been compromised but
the far more salient factor is Black’s superiority in
the center.)
11...Rd8= White cannot prevent ...d6-d5,
which as a rule (at least) equalizes for Black.;
12
9.Re1 Be6 10.Bf1 d5 11.Nc5
(11.exd5 Nxd5 12.Ne4 b6=)
11...dxe4 12.Nxe6 Qxd1 13.Rxd1 fxe6
14.Bc4 Kf7³ Black’s tripled pawns are not to be
scoffed at; they restrict White’s pieces by
controlling important central squares.
9...Be6 10.Kh1
White signals his intention to continue with
f2-f4. Since White’s bishop is on g5, it is no
longer realistic for Black to aim for ...d6-d5, so he
refocuses attention toward the queenside.
10.f4?! b5!„
10...Rc8 11.f4 a6
Black can take his time on the queenside
because if White continues with f4-f5 then Black
will happily take possession of the e5 square with
his knight.
12.Qe1
12.f5 Bd7! 13.Qc1 Ne5 14.Qf4 b5„;
12.Bf3 Nd7!? Just one idea of many.
13.Rb1 Bxc3 14.bxc3 Qc7³;
12.a4 Na5 13.Nxa5 Qxa5 14.Bd3 Rfe8
15.Qe2 Qb4 16.a5 Bg4 17.Qd2 Bd7 Black
threatens ...d6-d5 thanks to the pin on the
c3-knight. 18.Qe2?! Qxb2 19.Bxf6 Bxf6 20.Nd5
Bb5 21.Bxb5 Qxb5 22.Qxb5 axb5 23.Rab1 Rc5µ
0–1 (42) Anand,V (2715)-Topalov, V (2640)
Linares 1994.
12...b5
 13.f5 Bd7 14.Qh4 Ne5 15.Nd4 Rc5!„

Black intends ...Qd8-a8 with prospects for

the thematic ...Rxc3 Sicilian exchange sacrifice.
15...Nc4?! 16.Nd5 Nxd5 17.exd5ƒ

13
 2 (10.Nxc6?! bxc6 11.Qxc6 Bd7³) 10...Bd7„)
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6 6.Be3 Nf6
7.Be2
7.Nxc6 bxc6 8.e5 Ng8! (8...Nd5 I never
liked this pawn sacrifice but several grandmasters
have played it. 9.Nxd5 cxd5 10.Qxd5 Rb8∞) 9.f4
Nh6 10.Qd2 0-0 11.0-0-0 d6!=;
7.f4
7...0-0 8.Be2 Black can of course continue
with 8...d6 here and transpose to the ordinary
Dragon but I suggest we only do so after White
has displaced his knight from the active d4-square.
(8.e5 White can’t afford such extensions
before castling and completing development.
8...Ne8 9.Qf3 (9.Nxc6?! bxc6 10.h4?! d6 11.h5
Qa5 12.hxg6 hxg6µ; 9.Qd2 d6„) 9...d6 10.0-0-0
14
8...e5!
9.Ndb5 Black has safe paths to equality
here like 9...Ne8, but the following opportunity,
while messy and slightly risky, is far too appealing
to pass up.
(9.fxe5 Nxe5 10.0-0 d6 11.Bg5 h6 12.Bh4
g5 13.Bg3 Qb6 14.Kh1 Neg4 15.Qd3 Nh5„;
9.Nxc6 dxc6 10.fxe5 Ng4!„)
9...Nxe4!!N 10.Nxe4 d5 11.Nf2
(11.Ned6 a6 12.Nxc8 Rxc8 13.Na7 Nxa7
14.Bxa7 b6µ)
11...a6 12.Na3 exf4 13.Bc1
Black’s compensation is of a long-term
nature. For the sacrificed piece he currently has
two pawns, control over the center, and most of
White’s pieces are awkwardly placed. Black can
‘just play’ the position, even the computer gives its
approval (“0.00”) to several moves (13...Re8,
13...Qh4, 13...Nd4). The following is just one
possible continuation. 13...f3!? 14.Bxf3
 (14.gxf3?! Qh4 15.0-0 Bd4 16.c3 Ba7
17.Nc2 Bh3 18.Be3 Bxe3 19.Nxe3 Qg5+ 20.Kh1
Bxf1 21.Nxf1 Rfe8=)
14...Re8+ 15.Kf1 Nd4 16.c3 Nxf3
exceptions to our rule of thumb heuristics.
Nevertheless, Black is not better off here than he is
in the main line (7...0-0), it is just another path to
easy equality.
8.Bb5

8.exd5 Nxd5 9.Nxc6 bxc6 10.Nxd5 Qxd5

11.0-0 0-0= We have transposed to the 7...0-0
8.0-0 d5 line.
8...0-0 9.Bxc6

9.Nxc6 bxc6 10.Bxc6 Rb8©

9...bxc6 10.Nxc6 Qc7 11.exd5 e6!

17.gxf3
(17.Qxf3 d4! 18.Bd2 Be6 19.Re1 Qb6µ)
17...d4! 18.Bf4
(18.cxd4?! Bf5 19.d5 Rc8 20.h4 b5 21.Nb1
b4!µ)
18...dxc3 19.Qxd8 Rxd8 20.bxc3 Bxc3
21.Rd1 Be6 22.Rxd8+ Rxd8 23.Nb1 Bd4 24.a3
Rc8 25.Kg2 Rc2 26.Nd2 Ra2=
7...d5!?

12.0-0

12.Nb5 Qb7 13.Nd6 Qc7 14.Nb5 Qb7=;

12.Qd2 exd5 13.Nd4 Ne4 14.Nxe4 dxe4=
12...exd5 13.Nd4

13.Nb4 Ng4 14.g3 Nxe3 15.fxe3 d4³

13...Ng4 14.g3 Nxe3 15.fxe3 Qb7=

Black is not supposed to be able to play

this; the general rule is that, except in
extraordinary circumstances, Black must castle
before playing this. Obedience to that rule is likely
why this move has never been played by a 2400+
player (at least according to the database).
Computers are helpful for showing us the
15
16
 3

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.Nc3 Nc6 6.Be3 Nf6 7.Be2 0-0

8.Nb3
We previously considered this move
coupled with Bc1–g5; here Black proceeds in
essentially the same way.
8.Qd2?! d5 9.exd5 Nxd5 10.Nxd5
(10.0-0-0 Nxe3 11.Nxc6 Qxd2+ 12.Rxd2
Nf5 13.Nb4 Bh6–+ 0–1 (13) Salimbagat,R
(2269)-Panjwani,R (2393) World Open 2016)
10...Nxd4! 11.Nxe7+
(11.Bxd4 Qxd5 12.Bxg7 Qxg2! 13.Bxf8
Qxh1+ 14.Bf1 Qe4+ 15.Be2 Kxf8µ)
11...Qxe7 12.Bxd4 Bxd4 13.Qxd4 Re8
14.Qe3 Qb4+ 15.Qd2?!
(15.c3 Qa4µ)
15...Rxe2+ 16.Kxe2 Bg4+ 17.f3 Re8+
18.Kd1 Qxb2 19.Rc1 Qb6–+
8...d6!
11.Bf3
(11.d6 Bf5 12.Nd4 (12.dxe7 Qxe7 13.Nd4
Rfd8µ) 12...Qxd6=)
11...b6N
a) 11...Bg4 12.Bxg4 Nxg4 13.Qxg4
(13.Bd4 Bxd4 14.Nxd4 Nf6=) 13...Nxc2+
14.Ke2± (14.Kd2±);
b) 11...Bf5 12.Nd4 Nfxd5 13.Nxf5 Nxc3
14.Qxd8 Rfxd8 15.Nxe7+ Kf8 16.0-0 Nxc2
17.Bc5 Bd4 18.Nc6+ Bxc5 19.Nxd8 Rxd8
20.Rac1 Ne3 21.Rfe1!±;
12.d6 Bg4 13.Bxg4 Nxg4 14.Qxg4 Nxc2+
15.Ke2 Nxa1 16.Rxa1 Qxd6 17.Rd1 Qxh2 18.Bf4
f5 19.Qf3 Qh4 20.Nd4 e5 21.Bg3±
9.0-0
9.f4 Be6 10.g4

As per our recipe, we transpose to the

Classical Dragon when White commits to Nb3.
8...a5 I used to follow Donaldson and
Silman in playing this Accelerated Dragon-style
move, but unfortunately I don’t think it equalizes.
9.a4 d5?!
(9...Nb4?! 10.f4 (10.0-0 d5 11.e5 Ne4 This is the old ‘Rabinovich Attack’,
12.f3 Nxc3 13.bxc3 Nc6 14.f4 f6 15.exf6 Bxf6 popularized in the 1930s by Kan, Levenfish,
16.Qd2 e6∞) 10...d5 11.e5 Ne4 12.Nb5!²) Bondarevsky, and Alekhine. 10...Rc8
10.exd5 Nb4
17
 (10...d5?! 11.f5 Bc8 12.exd5 Nb4 13.Qd2
(13.Bf3 gxf5 14.a3 fxg4 15.Bg2 Na6 16.Qd3 e6
17.0-0-0 Nxd5 18.h3 g3 19.Rhg1 Qd6 20.Bxd5
exd5 21.Nxd5 Kh8 22.Bf4+– 1–0 (38)
Fischer,R-Reshevsky,S Los Angeles 1961; 13.d6
Qxd6 14.Bc5 Qf4 15.Rf1 Qxh2 16.Bxb4 Nxg4
17.Bxg4 Qg3+ 18.Rf2 Qg1+ 19.Rf1 Qg3+ 20.Rf2
Qg1+ 1/2–1/2 (20) Alekhine, A-Botvinnik,M
Nottingham 1936) 13...Nfxd5 14.Nxd5 Qxd5
15.0-0-0! Qxd2+ 16.Bxd2²)
11.f5
(11.g5 Nd7 12.h4 Nc5! 13.Nxc5 (13.Bxc5
dxc5 14.Qxd8 Rfxd8 15.Nxc5 Nd4 16.Nxe6
Nxe6³) 13...dxc5 14.Qxd8 Rfxd8 15.Bxc5 Nd4
16.Bxd4 Bxd4 17.Bd3 Bxc3+ 18.bxc3 Rxc3³)
11...Bxb3! Thank you to Daniel
Vanheirzeele for informing me of this idea.
12.cxb3
(12.axb3 This is a worse version for White
than 12.cxb3 since the c2-pawn falls in the main
line. 12...d5! 13.exd5 (13.Nxd5 Nxe4µ) 13...Nb4
14.Bf3 (14.Bc4 a6³) 14...Nfxd5!! 15.Nxd5
(15.Bxd5 Bxc3+ 16.bxc3 Rxc3 17.Bh6 Qb6!!
18.fxg6 hxg6 19.Be4 Rd8 20.Qe2 Qf6 21.Ra5
Qh4+ 22.Qf2 Qxg4–+) 15...Nxc2+ 16.Kf2 Nxa1
17.Qxa1 e6 18.f6 exd5 19.fxg7 Re8µ)
12...d5! 13.exd5 Nb4
18
14.Bf3
(14.Bc4 Here I prefer the simple 14...Qa5
but 14...b5 leads to interesting complications.
Unlike in the 12.axb3 line, 14...a6 is not playable
here because White has a2-a3 available to him in
this line. 14...Qa5!
a) 14...a6 15.g5 Ne8 16.a3²;
b) 14...b5 15.Nxb5 Nfxd5 (15...Nbxd5
16.Bxa7 Qa5+ 17.Qd2 Qxd2+ 18.Kxd2 Nxg4
19.Ke2²) 16.Bxa7 Bxb2 17.0-0∞;
15.0-0 Rcd8=)
14...Nfxd5!! 15.Nxd5
(15.Bxd5 Bxc3+ 16.bxc3 Rxc3 17.Bxf7+
Kxf7µ)
15...Nc2+ 16.Kf2
(16.Kf1 Qxd5! 17.Bxd5 Nxe3+³)
16...Nxa1 17.Qxa1 e6 18.f6 exd5 19.fxg7
Re8³ Practically speaking, White is just lost here
— computer level accuracy is required to only
remain slightly worse.
9...Be6
 13...Rxc3!–+ Just thirteen moves and
White is toast.;
11.g4 Na5 Another thematic idea to keep
in mind. 12.f5

10.f4

10.Qd2 d5=;
10.Nd4 d5 11.Nxe6
(11.exd5 Bxd5=) 12...Bc4!
11...fxe6 12.exd5 exd5= (12...Bd7?! We were happy to move back
10...Rc8 to d7 when the knight’s arrival on e5 was
imminent but here White’s attack is too fast.
13.Nd2 (13.g5? Rxc3!µ) 13...Nc6 14.Rf2 Ne5
15.g5 Ne8 16.h4‚)
13.g5
(13.Bd3 Nxb3 (13...Nd7 14.Bxa7∞)
14.axb3 Bxd3 15.cxd3 d5! 16.g5? d4µ)
13...Nd7

This position has been defended on the

Black side by such champions as Khalifman and
Kramnik. White’s attack is not to be
underestimated, but Black’s resilience and
counterattacking chances are fully adequate.
10...Qc8!? 11.h3 Rd8 12.Bf3 Nd7 13.Qd2
Nb6= 14.Bd3
11.Qe1 (14.Bxa7 Bxc3 (14...Be5!?©) 15.bxc3
Bxe2 16.Qxe2 Rxc3 17.Bd4 Rc8„)
11.f5?! White cannot just ‘go for it’; too 14...Bxc3! Black creates a target for his
many weaknesses are created in the process. queenside counterplay. The g7-bishop is often a
11...Bd7 12.g4 Ne5! 13.g5? liability anyway when White threatens f5-f6, etc.
(13.Nd2 Rxc3! 14.bxc3 Bc6©) 15.bxc3 Ne5 16.Bd4
19
 (16.Rc1 d5µ)
16...Nac6„;
11.Kh1 a6 12.g4
(12.Qe1 This transposes to 11.Qe1.; 12.Bf3
Nd7!„ Intending ...Nd7-b6-c4.)
12...d5! 13.f5 d4! 14.Nxd4 Nxd4 15.Bxd4
(15.Qxd4 Qxd4 16.Bxd4 Bc4³)
15...Bc4©

13...b4!N

13...Nd7 14.f5 Bc4

Black has full compensation, White has to
play accurately to not become worse;
11.Bf3
(14...Bxb3 15.cxb3² 0–1 (50)
Negi,P-Kryvoruchko,Y FIDE World Cup 2013)
15.Bxc4 bxc4 16.Nd2
(16.Nd4 Nxd4 17.Bxd4 Bxd4 18.Rxd4
Qb6 19.Rd1 Nf6 20.Rf3 Kg7=)
16...Bxc3
(16...Qa5!? 17.Nd5 (17.Nxc4 Bxc3
18.Qxc3 Qxc3 19.bxc3 Ncb8!³) 17...e6! 18.fxe6
(18.Nf4 Qxa2 19.fxe6 fxe6„) 18...fxe6 19.Nf4
Qxa2∞ (19...Nd8!?))
17.bxc3 Qa5 18.Nf3 Rb8„
14.Na4 Nxe4 15.Bb6 Qe8 16.Bxa6 Nf6
17.Bxc8 Qxc8

Black’s compensation is undeniable.

18.Nd4 Bd7 19.Nf3 Qb7 20.Bg1

11...Nd7!„ The reader might have noticed

this is a common way to meet Be2-f3. Black can
temporarily remove the d7-retreat square from his
bishop since White has relinquished control over
c4, and if Black achieves ...Nd7-b6-c4 he will take
over the initiative.
11...a6 12.Kh1

12.Rd1 Ng4„
12...b5 13.Rd1
20
 20...Nd8! 21.b3 Ne6 22.Nd4 Nd5 23.f5
Nef4 24.Qg3 Nh5 25.Qf3 Qa8©

21
 4 13.c4 d4 14.Bxb7 Rb8 15.Bd2 Rxb7 16.b4 e5=)
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6 6.Be3 Nf6 7.Be2 0-0 8.0-0 d5
If Black played 8...d6 instead, we would
have a position from the Classical Dragon where
Black would intend to play 9...d5 next. This is the
sense in which we are effectively up a tempo in
the Accelerated Dragon.
9.Nxc6
Not the most popular but in my opinion
White’s best.
9.exd5 Nxd5 10.Nxd5
The following choice is just a matter of
taste: both lead to very dry, equal endgames. I tend
to prefer 10...Nxd4 because it keeps Black’s pawn
structure in tact.
(10.Nxc6 bxc6 11.Nxd5 cxd5 12.Bf3 Bb7
22
10...Nxd4
(10...Qxd5 11.Bf3 Qc4 (11...Qa5 12.Nxc6
bxc6 13.Qc1 Rb8 14.c3 c5 15.Rd1 White is not
better, but our current world champion has taught
us with his games that enduring even mild
unpleasantness such as this can be made to be a
Herculean task.) 12.Nxc6 (12.Be2 Qb4³) 12...bxc6
13.c3 Bf5 14.Qb3 (14.Be2 Qe6 15.Qa4 Rfb8
16.Bc4 Qc8 17.Bb3 Be6=) 14...Qa6 15.Bc5 Rab8
16.Qa3 Qxa3 17.Bxa3 Rfc8=)
11.Bxd4
(11.Bc4 e5 12.c3 Be6=)
11...Qxd5 12.Bxg7 Qxd1 13.Raxd1 Kxg7
14.Rd2
(14.Bf3 Be6=)
14...Be6 15.Rfd1 Rfc8 16.a3 Rab8=
9...bxc6 10.e5
This system is underrated, and Black needs
to be a little careful here. Computers initially think
every move is absolutely equal, but they are slow
at grasping the positional pressure White can apply
if Black does not act fast.
10...Ne4!
Liquidating the position is in my judgment
the easiest path to equality for Black.
10...Ne8 This (like 10...Nd7) is thematic
but Black needs to play very energetically to cope
with his static weaknesses and the requisite level
of accuracy makes this a risky variation to enter
into. 11.f4 f6 12.exf6 Bxf6 13.Qd2 Rb8
(13...Nd6? 14.Nxd5±)
14.b3 Nd6 15.Rad1 Qc7
 (15...Bf5 16.Bf3 I prefer White.) 15...f5 16.b3
16.Na4 e5 17.fxe5 Ne4 18.Qe1 Bxe5
19.Rxf8+ Kxf8 20.Qh4 Kg8∞; 16.exf6 Bxf6 17.Bxf6 exf6=
10...Nd7 11.f4 e6 16...Kf7=
(11...f6?! 12.Nxd5! fxe5 13.fxe5 Nxe5
14.Nf4²)
12.Na4 a5
(12...f6 13.exf6 Qxf6 14.c3 These positions
are just much easier to play for White, so I
recommend avoiding them, even though Black
must objectively be fine.; 12...Qa5 13.c4 Ba6
14.Bd2 Qc7 15.Rc1 Qb7 16.Qe1 Rfc8 17.Rf3 dxc4
18.Qh4 Qb5 19.b3 Qd5 20.Be3 Qa5 21.Bxc4 Bxc4
22.Rxc4 Nb6 23.Nxb6 axb6³ 0–1 (40)
Paiva,J-Panno,O (2570) Sao Paulo 1972)
13.Qd2?!
(¹13.c4! f6 14.cxd5 cxd5 15.exf6²)
13...Ba6 14.c4 Qb8 15.cxd5 Bxe2 16.Qxe2
cxd5 17.Rac1 g5!?∞ 0–1 (38) Mahia,G
(2380)-Sorokin,M (2490) Pehuajo 1993. Neither side can make progress; Black
11.Nxe4 dxe4 12.Qxd8 should resort to waiting moves (...Bh6-g7, etc.)
until White initiates exchanges.
12.Qd4 Qd5=
12...Rxd8 13.Rfd1 Be6 14.Bd4

14...c5!

14...Rd7 15.a4 Nigel Short demonstrates

what can go wrong for Black if he isn’t careful.
15...f5?! 16.exf6 exf6 17.a5 Kf7 18.f3 exf3
19.Bxf3 Rc7 20.b4!² 1–0 (59) Short,N
(2685)-Yrjola,J (2485) Manila 1992.
15.Bc3

15.Bxc5 Rdc8=
23
 CHAPTER 2
7.Bc4: ANTI-YUGOSLAV
VARIATION
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6 6.Be3 Nf6 7.Bc4 Qa5
Contents
1. 8.sidelines & 8.0-0 0-0 9.sidelines
2. 8.0-0 0-0 9.Bb3 d6 10.sidelines
3. 8.0-0 0-0 9.Bb3 d6 10.h3 Bd7
11.sidelines
4. 8.0-0 0-0 9.Bb3 d6 10.h3 Bd7 11.Re1
5. 8.0-0 0-0 9.Bb3 d6 10.h3 Bd7 11.f4
1 and very often Black’s maneuvers in this system
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 Qa5
24
I follow Donaldson and Silman in referring
to this system as the ‘Anti-Yugoslav variation’. In
their words, “We call this system the
Anti-Yugoslav because 7...Qa5 more or less forces
White to castle kingside and thus avoids the
dangers of the Yugoslav Attack in the Dragon.”
This concession from White is not without
compromise from Black. In his repertoire book on
the White side of 1.e4, Negi points out the defects
in Black’s position: “...the queen on a5 is badly
placed — it just gets in the way of Black’s typical
plans like ...Na5, or ...b5/...a5. Now ...b5 can
always be met by a2-a3 without any worries.
Black also has the typical plan of ...Nxd4 and
...Bc6 available but if White avoids playing f2-f4,
the positions after Nd5 (intending to meet ...Bxd5
with exd5) seem quite pleasant for him. Moreover,
it is not easy for Black to find a new square for the
queen. Going to c7 will only encourage Bg5, with
ideas of Bxf6 and Nd5, so the most common plan
is ...Qh5. Although this may offer a few tactical
tricks, it’s hard to believe that the queen can be
well placed there.” I do not disagree with Negi —
Black’s queen usually does need to be relocated
from a5, because the c6-Knight wants that square,
are (in his words) “hard to believe”. That being
said, the fact that White players will be
incredulous towards Black’s maneuvers is a
positive feature of Black’s strategy (in this specific
variation and more broadly in the Accelerated
Dragon), not a reason to avoid this line.
In my experience playing this system over
the board for more than sixteen years, White
players have real difficulty managing their
position, because Black’s Queen on a5 means that GM Pilnik, who felt that White had enough
play neither resembles the Yugoslav Attack nor compensation for a pawn. Plausible as it may have
does it follow the usual paths in the Bc4/0-0 been at the time, modern computers demonstrate
Dragon. White therefore must be extremely concretely that White just ends up suffering here
accurate in order to pose any problems whatsoever — and being down material. 9...Qxb2 10.Nb5 Qe5
for Black, and more often than not White players 11.f4 Qb8 12.e5 Nxe5! 13.fxe5 Qxe5 14.Nf1 Ng4!
are unprepared to do so. Black, in his favor, is 15.Qxg4 Qxa1+ 16.Kf2 0-0µ)
usually confined to deciding between a few typical
themes, and that makes his life much easier.
In the introduction to the book I mentioned
that I was initially drawn to this system for its
incredible “cheapo potential”, but to reassure the
reader that this is a serious, safe, and reliable
counter to 7.Bc4, I point out that Ivanchuk relied
on this system at the highest stage when he
employed it against Grischuk in the London
Candidates, 2013 (see 10.Nd5!? in the next
subchapter).
8.0-0

Conventional wisdom dictates that this is

basically forced, but in my opinion 8.f3 is not to 9...Nxe4! 10.Bxe4 Bxc3+ 11.bxc3 Qxe4µ;
be underestimated either. 8.f3!? This move is highly underrated; for
8.Qd2? For White players who have never example, Donaldson and Silman, as well as Negi
seen 7...Qa5, this blunder is not a rare occurrence. give the move a question mark. 8...Qb4!
8...Nxe4! 9.Nxc6
(9.Nxe4 Qxd2+ 10.Nxd2 Nxd4µ)
9...Qxc3!!µ

Black must not allow White to play Qd2

I’ve had the pleasure of playing this over
the board many times. Black is up at least a pawn
in all variations.;
8.Nb3? Another frequent sighting when
facing unprepared opposition. 8...Qb4! 9.Bd3
(9.Nd2 Apparently, according to
Donaldson and Silman, this was an idea of the late
25
and 0-0-0, establishing a Yugoslav setup. The
‘Anti-Yugoslav variation’ wouldn’t be aptly
named if it allowed this! 9.Ndb5! White takes
advantage of Black’s queen being on b4, and
therefore the c7 square being vulnerable. This
move has surprisingly been overlooked by other
commentators.
(9.Bb3? Nxe4 10.Nxc6 Bxc3+ 11.bxc3
Qxc3+ 12.Ke2 dxc6 13.Qd4 Qxd4 14.Bxd4 Nf6
15.Rhe1 Rg8! Keeping the king centralized in
anticipation of the endgame. 16.Kf2 Nd5µ)
9...Qxc4!
(9...0-0 10.Bb3 d6 11.Qd2 Bd7 12.0-0-0
This is ‘just a game’, of course, but White has
succeeded in establishing the Yugoslav setup, so
on principle we do not allow it.)
10.Nc7+ Kd8 11.Nxa8 b6 12.a4 Bb7
13.Nxb6
(13.a5 bxa5 14.e5 Ne8 15.Ra4 Qe6 16.f4
Bxa8–+ 0–1 (31) Tatekhin,V (2154)-Danin,A
(2543) Belgorod 2011)
13...axb6 14.Bxb6+ Kc8 15.a5

By far the most popular move for White

here is 9.Bb3, which will be the subject of the next
few chapters. Here we consider alternatives for
White.
9.Nb3

This is a clever line: if Black continues

9...Qa5-d8, White intends to accept the loss of a
tempo with Bc4-e2, because Black has lost two
tempi with ...Qd8-a5-d8. Thus, a position from the
Classical Dragon is reached a full tempo up for
White. This was reason enough for Donaldson and
Silman to attribute a “?” to 9...Qd8, but in my
A highly unusual position has arisen. If judgment the Classical Dragon is ‘so equal’ that
memory serves, it was Bent Larsen who claimed even a tempo down, Black has no problems.
that in unusual positions the best moves are often 9.f3?! Qc5 10.Bb3
unusual themselves. The following maneuever
may appear unusual, but it will be a recurring
theme in this book. 15...Ne8! Black’s knight
unblocks the g7 bishop and heads to d6, targeting
the c4 square. 16.Na4 Nd6! 17.Be3 Qb5 18.Qd5
Ba6=
8...0-0

If you ask a seasoned Accelerated Dragon

player how many times the d4-knight is currently
under attack, they will surely say “three”, never
forgetting to include the g7-bishop, which can be
uncovered with tempo. 10...Ng4! 11.fxg4 Bxd4
26
12.Bxd4 Nxd4 13.Kh1 Qe5 14.Nd5 Nc6=;
9.Nd5 White cannot hope for an advantage
by releasing the central tension so quickly.
9...Nxd5 10.exd5 Nxd4 11.Bxd4 e5! 12.dxe6 dxe6
13.Bxg7 Kxg7=
9...Qd8!

I used to enjoy playing this when I was

younger. It amazed me that Black’s queen could
get away with such tightrope acrobatics.
Unfortunately I can’t recommend this for anything
other than, perhaps, blitz chess. 10.Be2 Qh4 11.f3
Nh5 12.Qd2
(12.Bf2?! One little inaccuracy from White
I recommend calling White out on their and Black is back in business. 12...Qg5 13.Nd5 e6
cleverness. If White plays the 7.Bc4 variation 14.Be3 Qh4 15.Qd2 Be5 16.f4 Bxb2 17.Rab1
against the Accelerated Dragon (as opposed to Bg7∞)
7.Be2) in the hope that they will be able to 12...Be5 13.f4 Bxc3 14.Qxc3 Nf6 15.Nd2
favorably transpose into a 7.Be2 line, then they d5 16.Nf3 Qh5 17.e5 Ne4 18.Qd3±;
probably don’t have much experience with Be2 9...Qc7
systems, so we should welcome this transposition.
Besides, it is no more correct to say that Black is
down a tempo in this line than it is to say that
Black is up a tempo in the Classical line as
compared to this one! Retreating the queen to d8 is
the simplest way for Black to handle the position;
he intends to continue in the same fashion as I
recommended in the Be2 chapter. It is worth
mentioning that super GM Malakhov has adopted
this move in more than one tournament game.
9...Qh5

10.Bg5² Perhaps it is a little harsh to

evaluate this as +=, because the truth is, it is just
an interesting game after, for example, 10...e6!?.
In this opening, though, what matters is the level
of counterplay, not objective assessment. With
White threatening Bg5xf6 and Nc3-d5, Black’s
counterplay is put on hold in order to defend
further, and so compared to 10...Qd8! which
27
genuinely offers equal chances, I assess this as
slightly preferable for White.
(10.f4 d6 11.Be2 a6 12.Qe1 Bg4 13.Bxg4
Nxg4 14.Nd5 Qd8 (14...Qb8 15.c3 e6 16.Nb6
Nxe3 17.Qxe3 Qa7 18.Rad1 Rad8=) 15.Bb6 Qd7
16.Rd1 Nf6 17.Nxf6+ Bxf6 18.e5² 1–0 (62) Wei
Yi (2641)-Bu Xiangzhi (2710) Zhongshan CHN
2014)
10.Be2
10.Bg5 d6 11.Re1
(11.Bxf6?! Bxf6³ Since Black’s queen is
on d8 (and not c7), Black can recapture with the
bishop without fearing Nc3-d5xf6.)
11...Nd7 12.Qd2 Nb6 13.Be2 Be6 14.Bh6
Rc8 15.Bxg7 Kxg7 16.Rad1 Nc4
(16...Ne5 17.Nd5 Bxd5 18.exd5 Qc7 19.c3
Rcd8 20.Nd4± 1/2–1/2 (56) Perez Candelario,M
(2474)-Malakhov,V (2664) Sanxenxo ESP 2004)
17.Bxc4 Bxc4 18.Nd5 b5= Black will
continue with a typical minority attack on the
queenside.
10...d6
18.Bb5 Qxd4+ 19.Rxd4 Nb6 20.Kf2 Rc5 21.Rd2
Rfc8³ 0–1 (58) Recuero Guerra,D
(2407)-Malakhov,V (2690) Benidorm ESP 2007.
11...Be6 12.g4
The test of whether Black can get away
with playing this system down a tempo is whether
such immediate attacks work for White.
Fortunately, they don’t, so as Black we should
provoke and welcome White to overreach. 12.Qe1
Rc8 13.Qh4 Nb4 14.Rac1? Rxc3! 15.bxc3 Nxa2µ
12...Rc8 13.g5
13.f5 Bd7 14.Qe1?! A typical method of
attack but simply inadequate here. 14...Ne5
15.Qh4?! Rxc3! 16.bxc3 Nxe4µ
13...Ne8 14.f5 Bd7 15.Qe1

White is overextended; the onus is now on

Black to hit back and demonstrate the soundness
of Black’s counterplay, even down a tempo.
15...Bxc3! 16.Qxc3

16.bxc3 Ne5³
Since this variation is discussed, tempo up, 16...Ne5 17.Qd2 gxf5! 18.exf5 f6!„
in the chapter on the Be2 (Classical) Accelerated
Dragon, here we will only consider attempts by
White to take advantage of the extra tempo with
aggressive play. If White sticks to slow
maneuvering, the extra tempo will be diluted over
time.
11.f4

11.Qd2 a5 12.a4 Be6 13.Rfd1 Rc8 14.f3

Nd7 15.Nd4 Nxd4 16.Bxd4 Bxd4+ 17.Qxd4 Qb6
28
 Black has counterattacking ideas such as
...Kh8/Rg8, ...Ne8-g7, and ...Bd7-c6. White’s king
is weaker than Black’s, because whereas Black’s
king can hide on h8, White’s corresponding h1
square is vulnerable to ...Bd7-c6 (a consequence of
White’s overextension).

29
 2
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 Qa5 8.0-0 0-0 9.Bb3
d6
White’s options are limited, as Black
threatens ...Nf6-g4.
10.Nd5
Played in Grischuk-Ivanchuk, London
Candidates 2013. After that game, I considered
this move to be the critical test of this entire
variation. However, I now consider it to be
harmless, and not just because of 10...Qd8!.
10.f3 This move is played often, but it is
clearly inferior to 10.h3, because Black can
continue in the same way as he does in the main
line, except White will eventually lose a tempo
when he inevitably plays f3-f4. 10...Bd7
(10...Nxd4 This is playable but
unnecessary. 11.Bxd4 Be6 12.f4 (12.Nd5 Bxd5
13.exd5 Nd7 14.c3 Bxd4+ 15.Qxd4 Rfe8 16.Rfe1
a6 17.Re2 Qc5 18.Qxc5 Nxc5 19.Rae1 Kf8
20.Bc2 b5=) 12...Qa6!? 13.Re1 Qc6 14.Nd5²)
11.Qd2 Nxd4 12.Bxd4 Bc6 13.Rad1 Nd7=
30
Black has solved his opening problems. It
is crucial to understand that if Black succeeds in
exchanging on d4, and exchanging queens with
...Qa5-c5, then Black is to be preferred (though not
enough to seriously plan for more than a draw), for
typical Sicilian reasons: more central pawns,
potential queenside minority attack, etc.;
10.Qd2? Nxd4! (10...Nxe4 11.Nxc6±)
11.Bxd4 (11.Qxd4 Ng4³) 11...Nxe4µ
10...Qd8!N
Black vacates the a5-square for his knight,
leaving White with the only piece in ‘no man’s
land’ (the 5th rank).
10...Nxd5!? 11.exd5 Ne5 12.h3 Qa6„;
10...Re8 11.Nxf6+ Bxf6 12.c3 Bd7
(12...Ne5?! This was Ivanchuk’s choice,
and it turned out okay for him — though he lost
the game, it was due to the clock rather than the
position. White, however, could have posed
serious problems with 13.f4! 13.f4! (13.h3 Qa6
14.Nc2 Nd3 15.Qf3 Be6 16.Bd4 Bxd4 17.Nxd4
Bc4= 1–0 (39) Grischuk,A (2764)-Ivanchuk,V
(2757) London ENG 2013) 13...Ng4 14.Bc1!±
Black’s knight is stranded on g4, and White’s
b3-bishop will coordinate with his f1–rook to
target f7 after f4-f5.)
13.f4 Nxd4 14.Bxd4 Bxd4+ 15.Qxd4 Qc5=
11.Nxf6+
11.f3 Black has his pick of ways to
equalise. 11...Bd7
(11...Na5 12.c3 Nxd5 13.exd5 Bd7=;
11...Nxd5 12.Bxd5 e6!? 13.Bxc6 bxc6 14.Nxc6
Qc7 15.Nd4 Rb8 16.Rb1 Ba6 17.Re1 Rfd8© With
two bishops and ...d6-d5 imminent, Black’s
compensation is undeniable.)
12.c3 Na5= Black’s queenside counterplay,
combining a minority attack with infiltration via
...Na5-c4, is simple and strong.
11...Bxf6
Be5!=
13...Bd7
Black prepares to take on b3, then charge
forward on the queenside with
...a7-a5/b7-b5/a5-a4. It is unclear what White is
doing; 14.f4 only creates weaknesses.
14.f4 Nxb3 15.axb3 e5! 16.Ne2 Bc6
17.Ng3 d5!=

12.c3

12.Nxc6 bxc6 13.c3 c5 14.Bc4 Rb8 15.Rb1

Qa5„ Black prepares to exchange light squared
bishops with either ...c8-e6 or ...Bc8-a6, releasing
pressure from the f7-pawn, so that White’s f4-f5
loses force. The onus will then be on White to
defend against Black’s ensuing b-file pressure.;
12.f4 Na5 13.c3 Nxb3 14.axb3

14...a5! 15.Qf3 Bd7 16.f5 b5„ This is

typical of Black’s counterplay in this variation.
12...Na5! 13.Qd2

13.Bd5 e6 14.Bb3 a6 15.Nc2 b5 16.Bf4

31
 3

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6

5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 Qa5 8.0-0 0-0 9.Bb3
d6 10.h3 Bd7

11.Qd2? A surprisingly common blunder

— the database shows that even grandmasters
have fallen for this. 11...Nxd4

The first major branching point for this

variation. Black intends ...Nc6xd4, followed by
...Bd7-c6 and ...Nf6-d7. Ideally, Black will
exchange dark squared bishops on d4, as well as
queens (after ...Qa5-c5). Black seeks out an
endgame for obvious reasons: Black’s trumps
include a central pawn majority and the prospect
of a long term queenside minority attack coupled
with play along the half-open c-file. White is
banking on obtaining a middlegame initiative as
compensation for these long term disadvantages,
so White will try to avoid exchanges. The two
most sensible moves here are 11.Re1,
recommended by Negi, and the main line, 11.f4.
These two moves will be covered in the next two
subchapters.
10...Nxd4 11.Bxd4 Be6 12.Nd5²
11.Nf3
32
12.Bxd4?!
(12.Qxd4 Most strong players catch their
blunder and choose to accept a worse endgame
rather than lose a pawn. 12...Ng4 13.Qd5 Qxd5
14.Nxd5 Nxe3 15.Nxe7+ Kh8 16.fxe3 Rae8
17.Nd5 Bxb2 18.Rab1 Be5 (18...Bg7 19.Nf6 Bxf6
20.Rxf6= 1–0 (75) Medvegy,Z (2474)-Stolz,M
(2391) Balatonlelle 2005; 18...Ba3 19.Nf6 Re7
20.Nxd7 Rxd7 21.Bd5= 1/2–1/2 (29) Simacek,P
(2476)-Malinovsky,K (2367) Olomouc 2009)
19.Nf6 Rd8 20.Bd5 b6 21.Nxd7 Rxd7³)
12...Nxe4! 13.Qe3
(13.Nxe4 Qxd2 14.Nxd2 Bxd4µ)
13...Bxd4 14.Qxe4 Bxc3 15.bxc3 Qg5µ
0–1 (70) Hou Yifan (2584)-Bacrot,E (2721)
Villarrobledo ESP 2009;
11.Qd3 Black proceeds with the usual plan
of ...Nc6xd4, ...Bd7-c6, and ...Nf6-d7. 11...Nxd4
12.Bxd4 Bc6 13.Rae1 Nd7 14.Bxg7 Kxg7 15.f4
Qc5+ 16.Kh2 a5 17.a4 Rad8
18.Qg3 Nf6
(18...Qb4?! Misplacing the queen; Black
wants to be able to meet f4-f5 with ...Qc5-e5.
19.f5 Ne5 (It is not too late for Black to admit his
mistake. 19...Qc5!=) 20.f6+² 1/2–1/2 (58) Wei Yi
(2557)-Yu Yangyi (2657) Shenzhen CHN 2013)
19.f5
(19.e5 dxe5 20.fxe5 Nh5 21.Qf2 Qxf2
22.Rxf2 g5! 23.g3 Rd4!„)
19...Qe5=;
11.Qe1?!
A move once played by Carlsen. White
sets up a positional trap (see 11...Nxd4), but Black
has an easy path to equality. 11...Nxe4!
a) 11...Nxd4?! The usual plan doesn’t work
here. 12.Bxd4 Bc6?! 13.Nd5! Qxe1? (13...Qd8
14.Nxf6+ Bxf6 15.Bxf6 exf6 16.Rd1²) 14.Nxe7+
Kh8 15.Rfxe1+–;
b) 11...Rfe8 Perfectly playable but
33
unnecessary given the strength of 11...Nxe4.
12.Rd1 Nxd4 13.Bxd4 Bc6 14.f4 (14.Bxf6 Bxf6
15.Nd5 Qc5 16.Nxf6+ (16.Nc7 Bxb2³) 16...exf6
17.Qd2 Rad8³) 14...Nd7 15.Bxg7 Kxg7 16.f5
(16.Kh2 Rac8 17.Qh4 h6 18.Nd5 e6 19.Nc3 Nf6
20.Qe1 1/2–1/2 (20) Nisipeanu,L
(2695)-Carlsen,M (2646) Sarajevo BIH 2006)
16...Nf6 17.Qh4 Qe5 18.Bd5 Rf8! (18...e6
19.Bxc6 bxc6 20.fxg6 fxg6 21.Rxd6 Qxd6
22.Qxf6+ Kg8 23.Qf7+ Kh8=) 19.Rf3 e6 20.Bxc6
bxc6=;
12.Nxc6
(12.Nxe4? Qxe1 13.Rfxe1 Nxd4µ)
12...Bxc6 13.Nxe4 Qxe1 14.Raxe1 Bxe4
15.Bc1 Bc6 16.Rxe7 Bf6 17.Re3 d5=
11...b5!
11...Rfc8 (This is the routine way we meet
White’s Nd4-f3, with the idea of ...Bd7-e8,
...Nf6-d7-c5. This is Negi’s recommendation for
Black, but in this particular move order I think we
can do better with 11...b5!.) 12.Ng5! Be8 13.f4²
White’s f1 Rook, b3 Bishop, and g5 Knight all
target f7, and Black doesn’t ideally want to play
...h7-h6 to kick out the g5 knight because then the
g6 pawn is weakened.
12.a3
12.Re1!? Black is no worse in the ensuing
complications. In my opinion, such lines aren’t
worth memorizing; Black’s position is
fundamentally sound (none of his pieces are bad)
so White’s pawn sacrifice is bound to be
inadequate — there are laws of justice governing
our beloved game. 12...b4 13.Nd5 Nxe4 14.Bf4
(14.Bc1 Nc5 15.Bg5 Bxb2 16.Rb1 Bc3
17.Nxc3 bxc3 18.Bxe7 Nxe7 19.Rxe7 Nxb3
20.axb3 Bc6=)
14...Nc5 15.Nxe7+ Nxe7 16.Bxd6
(16.Rxe7 Be6=)
16...Nf5 17.Bxc5 Qxc5 18.Qxd7 Rad8
19.Bxf7+ Kh8 20.Qb7 Bxb2 21.Rad1 Rxd1
22.Rxd1 Nd4=
12...Qa6!N
 With the intention of relocating the queen
to b7, where it simultaneously keeps one eye on
the center and one eye on the queenside minority
attack.
13.Re1

13.Qd2 Qb7 14.Bh6 Ne5 15.Nxe5 dxe5

16.Bxg7 Kxg7 17.Nd5 Bc6 18.Qc3 Nd7 19.Nxe7
Bxe4 20.f3 Rae8=
13...Qb7 14.Bg5 h6 15.Bh4 g5 16.Bg3
Na5=

34
 4 Nxd5 15.exd5 Ne5 16.Qd2 Rfe8 17.Bh6 Bf6
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 Qa5 8.0-0 0-0 9.Bb3
d6 10.h3 Bd7 11.Re1
18.Bg5 Bg7 19.Bh6 Bf6 1/2–1/2 (19) Brkic, A
(2573)-Mamedov,R (2653) Sarajevo BIH 2010;
b) 12.Nd5 Rfe8 13.Nxc6 bxc6 14.Nc3
Rcd8 15.Qd2 Bc8 16.Rad1 Nd7 17.Bd4 Nc5
18.Bxg7 Kxg7 19.Qe3 e5!= 0–1 (62) Mehar,C
(2246)-Kuzubov,Y (2626) Gurgon IND 2009;
12...Qh5
a) 12...Qc7?! 13.Bg5! Na5 14.Bxf6 exf6
15.Nd5² 1/2–1/2 (54) Petrosian,T
(2627)-Mamedov,R (2640) Bursa TUR 2010;
b) 12...Rfe8 13.Qd2 b5 14.a3 Qa6 15.Bg5
Qb7 16.Rad1 Na5 17.e5±

As I have mentioned, Black’s main plan in

this system begins with exchanging on d4 and
playing ...Bd7-c6. The prophylactic 11.Nf3!?
prevents this idea, but it is better for White to wait
a move to ask Black to commit a piece before
playing Nd4-f3, especially since on this move
11...Nxd4 is dubious. We see that one of the problems for Black
11...Rfc8 in this line is the d7-bishop; this is why I
recommend playing 11...Rfc8 and 12...Be8;
It was hard to decide which move to 13.Qe2
recommend in this line, but I am happy with this
choice for two reasons:
1) Black’s next few moves (except after
12.f4) can be pretty much played on autopilot
(...Bd7-e8, ...Nf6-d7-c5) — one shouldn’t
underestimate the value of such conveniences.
2) In the other lines, play often liquidates
into positions where Black is close to equal but
without many winning chances;
11...Rac8 (Even though I am not
recommending this, it will be instructive for the
reader to play through these variations to get a feel
for why 11...Rfc8 is a better choice, especially to
play for a win.) 12.Nf3
a) 12.Qe2!? Prophylaxis against ...Nc6-e5, Negi’s recommendation. He omits our next
because now White will be able to hit the e5 move though. 13...Ng4!N Black forces
knight with f2-f4, and the c4-square will be simplifications and the resulting position is close
unavailable to it. 12...a6 13.Rad1 Qc7 14.Nd5 to equal.
35
 a) 13...a6 14.Rad1 Ne5 15.Nxe5 Qxe5 12...Qh5!
16.f4 Qa5 17.Qf2 Rxc3 18.Bd2!± 1–0 (34)
Fedorov,A (2602)-Malakhov, V (2670) Warsaw
POL 2005;
b) 13...h6 14.Rad1 g5 15.Nd5 b6 16.Ba4!
A novelty found by Negi. The ensuing variation he
gives is not inspiring for Black. (16.c3 Nxd5
17.Bxd5 Ne5 18.Nxe5 Qxe2 19.Rxe2 Bxe5 20.h4
Bb5 21.Ree1 Bf6 22.hxg5 hxg5= 1/2–1/2 (50)
Wang Hao (2519)-Malakhov,V (2670) Khanty
Mansyisk RUS 2005) 16...Rfe8 17.c3 Nxe4
18.Bxb6 Nf6 19.Nxf6+ Bxf6 20.Bd4 Nxd4
21.Rxd4!²;
14.hxg4 Bxg4 15.Qb5
(15.Nd5 g5 16.Bxg5 Bxf3 (16...Ne5
17.Qe3 Nxf3+ 18.gxf3 Bxf3 19.Qxf3!! Qxf3 This move is a recurring theme in this
20.Re3 Qh5 21.Rg3 Rc5 22.Bxe7 Rxd5 23.Bxd5±) Nd4-f3 variation. Black intends to continue with
17.Qxf3 Qxg5=) ...h7-h6 and ...g6-g5, with counterplay against
15...Bxf3 16.Qxh5 Bxh5 17.f3 g5 18.Bxg5 White’s king. 13.Rb1!? Prophylactically defending
Bg6 19.Nd5 Bxb2 20.Bxe7 Bxa1 21.Rxa1 Nxe7 the b2-pawn which will be vulnerable after
22.Nxe7+ Kg7 23.Nxc8 Rxc8 24.Rd1 Rc6 25.Rd5 Nc3-d5.
f6 Black should be able to hold, but I recommend (13.Qe2 Ng4!
avoiding all this and playing 11...Rfc8 instead.;
11...Rad8!?

A stunning resource that we see more than

A move not mentioned by Negi, but
deserving of our attention if for no other reason
than the fact that Korchnoi played it in 2003. In
my judgment this is a serious alternative to our
main line, and perhaps the place to go if our main
line runs into theoretical problems. 12.Nf3! This
must be White’s best.
(12.Qd2 Nxd4 13.Bxd4 Bc6 14.Rad1 Rfe8
15.Qe3 Nd7 16.Bxg7 Kxg7 17.Kh1 Qc5= 1/2–1/2
(53) Sadvakasov,D (2523)-Korchnoi, V (2642)
Astana KAZ 2003)
36
once in this chapter. 14.hxg4 Bxg4 15.Qb5
(15.Nd5 g5! (15...Ne5? 16.Nf4+–) 16.Bxg5 Ne5
(16...Nd4? 17.Nxd4 Bxe2 18.Bxe7 Bxd4
19.Bf6!!±) 17.Qe3 Nxf3+ 18.gxf3 Bxf3 19.Bh4
Qg4+ 20.Bg3 Qh5 21.Bh4=) 15...Bxf3 (15...d5
Unfortunately White gets away after this. 16.exd5
Bxc3 17.bxc3 Bxf3 18.gxf3 Ne5 19.Kf1!!±)
16.Qxh5 Bxh5 17.f3 g5 18.Bxg5 Rd7= Black is
about equal.)
13...h6 14.Nd5 g5
(14...Nxe4? 15.Nf4 Qf5 16.g4 Qa5
17.Nxg6±)
 15.Bd4 Qg6 16.Bxf6 Bxf6 17.c3 13...h6!
(13...Ng4? This doesn’t quite work here,
because the f7-pawn’s weakness poses a problem.
14.hxg4 Bxg4 15.Qc4!±)
14.Rad1 g5!

17...Kh8!„ Black is ready for ...Rf8-g8 and

...g5-g4. White players won’t know what hit
them!;
11...Nxd4?! 12.Bxd4 Bc6
(12...Qg5 13.Be3²)
13.Nd5! Black is not given time to play
...Nf6-d7; he has to make a concession either by
allowing White to capture on f6, damaging
Black’s pawn structure, or by opening up White’s
e-file. 13...Bxd5 14.exd5 Rfe8 15.c4 a6 16.Bc3
Qc7 17.Rc1 Nh5 The following is a
recommendation of Negi’s. 18.Ba4!? Rec8
19.Qd2²
12.f4
15.Nd5 Nxd5 16.Bxd5 Bxb2 17.Rb1 Bc3
18.Rxb7 Bxh3 19.Nxg5 Qxe2 20.Rxe2 Bg4
21.Nf3 e6 22.Bxc6 Rxc6 23.Bxh6 We have been
following Carlsen-Radjabov, 2010 (1–0). Black’s
opening preparation was excellent, but his play
could have been improved here. 23...Rb6! 24.Rc7
Ba5 25.Rc4 e5!= Black’s two bishops fully
compensate for the pawn; if I had to choose I’d
take Black.;
12.Qd2

I am a little surprised that Negi didn’t

recommend this in his book. This rare move is
extremely underrated, and has only been played by
one 2500+ player — none other than Negi himself
(Negi-Khalifman 2007). The main upshot of this
move from White’s perspective is that it deters
Black from his usual, comfortable course of
...Bd7-e8 and ...Nf6-d7, because White’s f-file
pressure can become overwhelming.
12.Qe2 Qh5!
(12...Be8 is playable as well but 12...Qh5 is
better)
13.Nf3
a) 13.Nxc6 Bxc6 14.f3 d5! 15.g4 (15.Nxd5 12...Ne5
Nxd5 16.exd5 Bxd5 17.c3 Bc6=) 15...Qxh3 (12...Be8 As always Black can follow
16.exd5 Bd7 17.Qg2 Qxg2+ 18.Kxg2 b5!„; through with this plan, but with White’s queen on
b) 13.Qxh5 As usual Black welcomes the d2 Black has the added possibility of
trade of queens. 13...Nxh5 14.Rad1 Na5 15.Nd5 ...Nc6-e5-c4.)
Kf8=; 13.Bh6 Nc4 14.Bxc4 Bxh6 15.Qxh6 Rxc4
37
16.Nb3
a) 16.Rad1 Qb6 17.e5 dxe5 18.Nf3 e4
19.Ne5 Rd4 20.Rxd4 Qxd4 21.Nxd7 Qxd7
22.Nxe4 Nxe4 23.Rxe4 Qd1+ 24.Kh2 Qd6+
25.Kg1 Qd1+ 26.Kh2 Qd6+ 1/2–1/2 (26)
Akopian,V (2712)-Bacrot,E (2721) Jermuk ARM
2009;
b) 16.Nf3 Qh5 17.Qe3 Qc5=;
16...Qb6 17.Rad1 a5 18.e5 dxe5 19.Rxe5
a4 20.Nd2 Rd4 21.Qe3 Qd6= 1–0 (57) Akopian,V
(2696)-Bacrot, E (2728) Nalchik RUS 2009;
12.Nf3 It is a good sign for our 11...Rfc8
that 12.Nf3, which is supposed to be White’s main
idea after 11.Re1, is not very effective here.
12...Be8! 18.Bh4 Threatening Qd2-h6.
(18.Re3 Rc7!„; 18.c4 Qxd2 19.Rxd2 Kf8
20.f3 Nd7=)
18...Kg7 19.c4 Qxd2 20.Rxd2

13.Qd2
a) 13.Nd5 Nd7 (13...e6 Black should be
fine here as well. 14.Nxf6+ Bxf6 15.c3 Ne5
16.Nxe5 dxe5=; 13...Nxe4 14.Bf4 Nc5 15.Rxe7
Bxb2 16.Rb1 Bg7∞) 14.c3 e6 15.Nf4 Nc5
16.Qxd6 Nxe4 17.Qd1 Nc5 18.Bc2 Nd7=;
b) 13.Bg5 Nd7 14.Nd5 e6 15.Ne7+ Nxe7
16.Bxe7 Bxb2 17.Rb1 Bg7 18.Bxd6 Ne5 19.Bxe5
Bxe5 20.Bd5 Bc3 21.Bxb7 Bxe1 22.Qxe1 Qxa2
23.Bxc8 Rxc8 24.Ra1 Qxc2 25.Rxa7 Qc1 1/2–1/2
(25) Kurnosov,I (2602)-Bacrot, E (2722) Moscow
RUS 2009;
13...Nd7 14.Rad1
(14.Bh6 Bxc3 15.bxc3 Nce5 16.Nxe5 dxe5
17.Qg5 Nf6„ In addition to winning the c3-pawn,
White’s b3 Bishop will be vulnerable to Black’s
...a5-a4.)
14...Bxc3! 15.bxc3 Nce5 16.Nxe5 dxe5
17.Bg5 Nf6
20...Kf8! 21.f3
(21.Bg3 Nd7 22.Red1 Nb6 23.Bxe5 Nxc4
24.Bxc4 Rxc4³)
21...Nd7 22.Red1 Nc5³;
12.Qd3 Played by GM Timofeev, but as
best I can tell the Queen just becomes a target to
either ...Nf6-d7-c5 or simply ...Nc6-e5 12...Be8
(12...Ne5 13.Qe2 Qa6!=)
13.Rad1 Nd7 14.Nd5 Qd8!= 0–1 (43)
Timofeev,A (2657)-Malakhov, V (2690) Tomsk
RUS 2006;
12.Nd5 This is Negi’s recommendation. He
only considers 12...Nxd5 and 12...Qd8, but it
seems to me that Black’s best is 12...Re8!

38

I don’t see anything for White here; as a
general rule I tend to think that if White rushes
with Nc3-d5 then as long as Black isn’t obligated
to take it, he should be fine (this doesn’t mean that
if he is forced to take it then he is necessary not
fine!).
(12...Nxd5 13.exd5 Ne5 14.Bg5 Re8 15.c3²
(Negi); 12...Qd8 13.Bg5! Nxd4 14.Bxf6! Nc6
15.Bh4 Bxb2 16.Rb1 Bg7 17.Ba4± (Negi))
13.Nxf6+
a) 13.Nxc6 bxc6 14.Bd2 Qd8 15.Nc3 c5
Black has no problems. If White starts to
get ambitious then Black come out on top. 16.e5?!
dxe5 17.Rxe5 c4 18.Bxc4? Qc7µ;
b) 13.Bd2 Qc5 14.Nxc6 (14.Nf3 Nxd5
15.exd5 Na5 16.Be3 Qb5=) 14...Nxd5!
(14...bxc6?? 15.Nc7+–) 15.exd5 bxc6 16.Be3
Qb5=;
13...Bxf6 14.c3 Qc7! Vacating the a5
square for the knight.
(14...Nxd4!? This is acceptable but
39
unnecessary. 15.Bxd4 (15.cxd4 e6!=) 15...Bxd4
16.Qxd4 (16.cxd4 e6!=) 16...Qc5 17.Qd2 Bc6
18.Rad1 Rad8=)
15.Nb5 (15.f4 Na5=) 15...Qb8 16.a4 a6
17.Nd4 Na5 18.Ba2 Qc7=
12...Nh5!N
In this particular variation Black is well
advised to abandon the ...Be8/ ...Nd7 plan, because
as we see in the alternatives White is able to build
up pressure on f7 after f4-f5/Re1–f1/ Qd1–f3.
Instead, we take immediate advantage of the
weakness White has created: Black threatens to
win the f4 pawn, so White’s options are limited.
12...Nxd4 13.Bxd4 Bc6 14.Nd5
14...Bxd5
(14...Re8 15.c3 Nd7 (15...Nxd5 16.exd5
Bxd4+ 17.Qxd4 Bd7 18.Kh2²) 16.f5! (16.Bxg7
Kxg7 17.Qd4+ e5 18.Qf2 Bxd5 19.Bxd5 Qc5
20.Qxc5 dxc5 21.Rad1 (21.Bxb7 Rab8 22.Bc6
Re7=) 21...Re7 22.f5 Nf6=) 16...Bxd4+ 17.cxd4
Bxd5 18.Bxd5 Nf6 19.fxg6 hxg6 20.Bb3²)
15.exd5 Re8 16.c3
(16.c4 a6 17.Bc3 Qc5+ 18.Bd4 Qa5
19.Bc3 Qc5+ 1/2–1/2 (19) Negi,P
(2529)-Khalifman,A (2632) Amsterdam NED
2007)
16...b5 17.a3 Nd7 18.Bf2²;
12...Be8 Unfortunately Black can’t stick to
the routine and play this move here; White builds
up too much pressure. 13.Qf3 Nd7 14.Rad1 Nc5
15.Qf2 Nxb3 16.axb3 Nb4

13.f5 Nf6 Having provoked White into

weakening the e5-square, Black retreats to f6,
guarding d5 against Nc3-d5, and prepares
...Nc6-e5.
(13...Nxd4 Black is perfectly resilient
against White’s attack here as well. 14.Bxd4 Bc6
15.Rf1 Rf8 16.g4 Nf6 17.Nd5 Bxd5 18.exd5
Nd7=; 13...Ne5?! 14.Nd5²)
14.Rf1
(14.Nxc6 bxc6 15.g4 Be8!=)
14...Ne5
This was played by Bacrot in Burg-Bacrot,
2011 where White’s play could have now been
improved upon.
(16...Nxd4 17.Bxd4 Bxd4 18.Rxd4
(18.Qxd4 Qc5=) 18...Bc6 19.b4! Qb6 20.b5! The
beginning of a long forcing variation which is
unequivocally in White’s favor. 20...Bxb5 21.Nd5
Qd8 22.Qh4 Rc4 23.Nxe7+ Kf8 24.Rxc4 Bxc4
25.Qf6! Ke8 26.Nd5 Qxf6 27.Nxf6+ Ke7 28.Nxh7
f6 29.e5 dxe5 30.fxe5 f5 31.Ng5²)
17.f5!± White will simply pile up on the
f-file with Re1–f1, and I don’t see a constructive
plan for Black.
13.Nde2
A typical Dragon position; Black is
salivating at the prospect of an exchange sacrifice
on c3. 15.Qe2
(15.Nd5 Nxd5 16.Bxd5 Bc6 17.Nxc6 bxc6
18.Bb3 Rab8=)
15...Rxc3! 16.Bd2
(16.bxc3 Nxe4³)
16...Rxc2!
(16...Bb5 17.Nxb5 Qb6+ 18.Kh2 Rc5
19.a4 (19.Nd4 Rac8„) 19...Rf8 20.Rae1 a6 21.Nc3
40
Rcc8 22.g4²) (29.h4 d5=)
17.Bxc2 29...g5! 30.Bd7
(17.Nxc2 Qb5©) (30.Be4+ Kh6=)
17...Qb6 18.Bc3 30...Kg6 31.Bxe6 Nxe6 32.Rd7 Rb6=;
(18.Be3 Qxb2 19.Rab1 Qxa2 20.Rxb7 Bc8 13.Nf5 Thanks to my editor, GM Romain
21.Rxe7 Kf8 22.Rc7 Qa5 23.Rxc8+ Rxc8³) Edouard, for assuaging my concerns about this
18...Nc6 19.Qf2 variation. In his words, “White better give mate, or
(19.Rad1 Ng4 20.Kh1 Nxd4 21.Bxd4 Bxd4 he is worse!” A more modest assessment is that
22.hxg4 Rc8 23.b3 g5=) Black has at least equal chances in the ensuing
19...Nxe4! 20.Bxe4 Bxd4 21.Bxd4 Qxd4 complications. 13...Bxf5 14.exf5 Nf6 15.fxg6
22.Qxd4 Nxd4 hxg6

23.f6
(23.fxg6 hxg6 24.Bxb7 Rb8 25.Bd5 e6
26.Bb3 a5 27.Rad1 Nxb3 28.axb3 d5 29.Ra1 Rxb3
30.Rxa5 Rxb2 31.Ra7 Be8 32.Ra8 Kf8 33.Rc1
Rb7 34.Rcc8 Re7=)
23...e5
(23...Bc6 24.Bxc6 bxc6 25.fxe7±)
24.Bxb7 Rb8 25.Bf3 a5 26.b3 h5 27.Rac1
Be6 28.Rc7 Kh7
16.Qd3
(16.g4 Rd8 17.f5 d5 18.g5 Nh5 19.fxg6
fxg6 20.Bf2 Kh7„)
16...Rd8! This move saves the day. Black
indirectly defends the g6 pawn. 17.g4 (17.Qxg6?
d5!µ) 17...e6 18.Rad1 (18.f5 gxf5 19.gxf5 Qxf5
20.Qxf5 exf5=) 18...d5 19.Bf2 Ne7= Preventing
f4-f5. Things are at a standstill; it is hard to see
how either side makes progress, but Black is
certainly no worse.;
13.Nd5 Re8 Black maintains his threat to
win the f4 pawn, as its defender on d5 can be hit
with ...e7-e6. 14.g4 Nxd4 15.Bxd4 Nxf4 16.Nxf4
e5 17.Nxg6 exd4 18.Nf4 Qg5 19.Qf3 Be5 20.Rf1
Be6=
13...Qd8!

After 13...Qc7 14.Qd2 Na5 15.Rad1 we

would like to play ...Bd7-e8 but White then has
Nc3-d5xe7. This is why we play 13...Qd8!
15...Be8? 16.Nd5 Qd8 17.Nxe7+ Qxe7 18.Qxa5±
14.Qd2 Na5 15.Rad1

29.Bc6 15.Nd4?! Nc4 16.Bxc4 Rxc4³

41
 15...Be8 25.Nb4 Rcc8 26.Qh4
(26.Nd5=)
15...b5 16.e5² 26...Qg7 27.Red1 g5 28.Qf2 g4! 29.hxg4
16.f5 Nxb3 17.cxb3 Bc6 18.Bd4 Nf6 Nf6 30.Qh4 Nxg4„;
19.Nf4 21.Nfd5 Bxd4 22.Qxd4 f6 23.Nf4 Ne5
24.Ne6 Qh6 25.Nd5 Bxd5 26.Qxd5 Rab8= White
has no productive discovered check, and Black’s
e5-knight is more valuable than his counterpart on
e6, which is hitting thin air. Computers call this
position 0.00 but I suspect most White players
would collapse rather quickly here in over the
board play — the squares Black controls are more
valuable than the squares White controls.
21...Qxg7 22.Nfd5 Bxd5 23.Nxd5 Qe5=

19...Qf8!

A nice little queen sortie; Black intends to

continue with ...Nf6-d7, and then recapture on g7
with his queen, potentially heading to e5
eventually.
20.Qe3 Nd7 21.Bxg7

21.Ncd5 Bxd4 22.Rxd4

(22.Qxd4 g5 23.Nd3 f6=

The weak e6-square is inaccessible to

White. As we will see, in other lines Black allows
White to plant a knight on e6; it turns out to not do
much from there.)
22...Kh8 23.Qg3 Bxd5 24.Nxd5 Rc2
42
 5

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6

5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 Qa5 8.0-0 0-0 9.Bb3
d6 10.h3 Bd7 11.f4

The most popular move, though I suspect it

may become overshadowed by 11.Re1, especially
given the popularity of Negi’s book which
recommends it.
11...Nxd4 12.Bxd4 Bc6

13...Nxe4! 14.Bxg7 Qc5+ 15.Kh1 Nf6!µ;

13.Qe2? The database shows that a
2478-rated player fell for this. 13...Qb4! 14.Rad1

Black proceeds with his usual plan. Next

he’d like to play ...Nf6-d7, and ideally exchange
dark squared bishops and queens.
13.Qd3
14...Nxe4!µ;
13.Qe1?! This 7...Qa5 line is one which 13.Nd5 Rae8!
induces even grandmasters to blunder within the
first 15 moves. 13...Qb4! 14.Bxf6
(14.Rd1 Nxe4!!µ)
14...Bxf6 15.Nd5 Qc5+ 16.Kh1 Bxd5
17.Bxd5 Qxc2 18.Rc1 Qxb2 19.Rb1 Qd4µ 1/2–1/2
(47) Hector,J (2551)-Efimenko,Z (2680)
Emsdetten GER 2008;
13.Qf3?

Abiding by the general rule that if we can

43
avoid taking on d5 then we should — here White’s
weak e4 pawn more than compensates for the
doubling of our f-pawns in the event that White
takes on f6. 14.f5
a) 14.Nxf6+ exf6³;
b) 14.Bxf6 Bxf6 15.Nxf6+ exf6 16.Qxd6
Rxe4 17.Rad1 (17.Qxf6?? Re2 18.Rf2 Rxf2
19.Kxf2 Qd2+–+) 17...Kg7µ;
c) 14.Qd3 Nd7 15.Rad1 Bxd4+ 16.Qxd4
Qc5=;
14...Bxd5 15.exd5 Nd7
(15...b5?! An odd recommendation of
Donaldson and Silman. This move seems
altogether wrong-headed. 16.c3²)
16.Kh1 Bxd4 17.Qxd4 Ne5 18.f6 Nd7!³ In Black is slightly better. 19.f5 Qe5 20.Rde1
the long run, White’s dark squared weaknesses b5 21.Nd5 Nh5 22.Re3 a5 23.f6+ Kh8 24.a3
will tell. Bb7–+ White’s initiative has run out of steam and
13...Nd7 Black is ready to collect the f6 and thereafter the
e4 pawn.
13...Rad8 This is the recommendation of (24...Qxb2 25.Ne7 Ba8 26.Qxb5 Nxf6–+
Donaldson and Silman, but their problems with 27.Qg5 Nxe4 28.Qh4 Qe5?? (¹28...g5 29.Qh6
13...Nd7 can be solved. Qg7–+) 29.Rxf7!± Morozevich,A
14.Bxg7 Kxg7 (2760)-Carlsen,M (2864) Moscow 2013);
15.Qd4+ Kg8 16.Kh1
(16.Rf2 Qc5 17.Qd3 b5 18.Bd5 Bxd5
19.Nxd5 e6 20.Nc3 Nf6„; 16.Rad1 Qc5=)
16...Qc5 17.Qd3 Nf6

15.Kh1

This used to be considered good for White,

because Black would follow up with 15...Nc5 and
further queenside play, rather than reverting back
to the kingside and solidifying control over the
center with ...e7-e6!
15.Nd5 Qc5+ 16.Kh1 e6 17.Nc3 Nf6
18.Rad1 Rad8
Black has ideas of ...Rad8, ...e7-e6, and
then either (or both) ...b7-b5/...a7-a5 or ...d7-d5.
The dark square weaknesses in White’s position,
especially g3, are a significant liability. 18.Rae1
e6 19.f5 Qe5 20.Nd5 Nh5 21.Ne7+ Kg7 22.Nxc6
bxc6³;
15.Rae1 Qc5+ 16.Kh2 e6!

44
 for later on) breathes new life into this variation.
15...Nc5 This is currently the most popular
move but I believe it too much neglects Black’s
kingside, as the following game illustrates.
16.Qd4+ Kg8 17.Rae1 Nxb3 18.axb3 Qc5 19.Qd2
Rad8 20.Nd5 Bxd5 21.exd5 Rc8 22.c3 Qb5 23.f5!

Typical for this line. Black tempts White’s

f4-f5, when either his knight or queen would
discover immortality on e5.
(16...b5?! 17.Bd5²)
17.Rd1 Nf6 18.Rfe1 Rad8³ The position
doesn’t look too bad for White, but all the
dynamic potential in the position lies in Black’s 23...Rc5 24.Qh6 Qd3 25.Rf4 Qg3 26.Ref1
hands: he can take his time expanding on the gxf5 27.Rxf5 f6 28.R1f4 Qe1+ 29.Kh2 Rf7
queenside, and break with ...d6-d5 at the right 30.Rg4+ Kh8 31.Qh5 1–0 (31) Guseinov,G
moment, while White can only wait in angst. (2573)-Ahmadinia,E (2195) Iran 2005.
15...Nf6!N 16.Rae1

16.Nd5!?

White seeks to prevent Black from playing

...e7-e6, but it can still be played after a bit of
preparation. 16...Qc5 17.Rae1 Rae8 18.a4

This idea of countering in the center and

kingside, instead of playing on the queenside (at
least for now — that is certainly on the to-do list
45

18...e6! 19.Nxf6 Kxf6 Black has nothing to
fear. White can’t target d6 without dropping e4
either. 20.f5 Kg7 21.f6+
(21.fxe6 fxe6 22.Rxf8 Kxf8 23.Qf3+ Kg7
24.Rf1 Qe5 25.Qf7+ Kh8 26.c3 Qxe4 27.Qf6+
Kg8 28.Qf7+ Kh8=)
21...Kh8= Do not fear ghosts: White has no
mating attack, and to divert his pieces to the
kingside in the hope of obtaining one would allow
Black to dominate the center more than he already
has.
16...e6!
17.e5
17.Qxd6? Rfd8 18.Qe7 Rd7 19.Qa3 Qxa3
20.bxa3 Rad8³ Black will win his pawn back after
...Rd4 (and if e4-e5 then ...Nf6-h5), and after he
does his superior pawn structure will yield him the
advantage.;
17.Qe3 Rad8 18.f5 Qc5
46
(18...Qe5 Black also comes out on top in
the ensuing complications here. 19.Qxa7 Nxe4
20.Nxe4 Bxe4 21.f6+ Kh8 22.Qe3 d5 As long as
Black can meet Qh6 with ...Rg8 (and Rf4 with
...g5) — which Black always can as long as his
king is on h8, there is nothing to worry about.
White’s b3-Bishop is missing in action. 23.Qh6?
Rg8 24.Rf4? g5–+)
19.fxe6 fxe6 20.Bxe6 Nxe4! 21.Rxf8 Rxf8
22.Qxc5
(22.Nxe4?! Rf1+! 23.Kh2 Qe5+ 24.g3
Rxe1 25.Qxe1 Qxe6 26.Qc3+ Kg8 27.Nf2 Qxa2µ)
22...Nxc5 23.Bd5 Rf2³
17...dxe5 18.fxe5 Nd7 19.Qe3
19.Qg3 Rad8„
19...Qc5 20.Qf4 Rad8
White needs to play perfectly to not be
worse here; in practice most players will collapse,
either allowing Black queenside counterplay or
central control.
Generally speaking, exchanges favor Black
(White’s e5 pawn will be weak in any endgame),
and Black should keep a lookout for counterplay
with ...f7-f6(f5). A sample variation may continue
as follows
21.Nd1! h6
21...f6!? 22.exf6+ Nxf6„
22.Nf2 f5! 23.exf6+
23.Bxe6?! g5 24.Qh2 f4 25.Bxd7 Rxd7
26.Nd3 Rxd3! 27.cxd3
 27...f3µ
23...Rxf6 24.Qg3 Rdf8 25.Rxe6 Bb5
26.Rxf6 Rxf6 27.c4 Bc6„

47
 CHAPTER 3
7.Bc4: MY SYSTEM
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.Nc3 Nc6 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6!?
Contents
1. 9.sidelines
2. 9.f3 Qc7 10.sidelines & 10.Qd2 b5
Appendix. 9.f3 Qc7 10.Qd2 Na5
1 work is undoubtedly incomplete. There are reams
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4
48
The Anti-Yugoslav Variation (7...Qa5)
seemed to me to be in trouble after
Grischuk-Ivanchuk, Candidates 2013 (recall
Grischuk’s 10.Nd5!?). In the summer of 2013 I
had not yet discovered the equalizing 10...Qd8!
novelty, and I found myself struggling to meet the
7.Bc4 variation. I searched the database for some
guidance, but to no avail. At the time I was
simultaneously a Taimanov Sicilian player as well,
and at some point it occurred to me that nobody
plays Bc4 against the Taimanov, presumably
because its scope on c4 is stunted by the e6 pawn.
This led to the idea of meeting the Bc4 variation
with an eventual ...e7-e6, and transporting ideas
from the Taimanov such as
...a6-...Qc7-...b5-...Bb7-...Nc6-a5(e5), and even
...Nf6-e8-d6. The process of developing this
system has been some of the most enjoyable chess
research I have engaged in, and I am pleased to
say it has resulted in a powerful and truly novel
system. As the ‘prime mover’ of this project, my
of unexplored possibilities, and I encourage
readers to attempt their own contributions. The
following chapters are by no means the final —
only the first — word on this system.
7...0-0 8.Bb3
Black’s next few moves can almost be
played automatically, irrespective of what White
does: ...a6, ...Qc7, ...b5, ...Bb7. Although our main
idea is to play ...e7-e6, the first priority is to
complete our development and connect the rooks.
8.f3?! White must not be permitted to
reach a Yugoslav Attack setup with Qd2 and
0-0-0. 8...Qb6!
9.Bb3
(9.a3 Qc5 (9...Nxe4?? 10.Nd5 Qa5+
11.b4+–) 10.Ba2 Ng4 11.fxg4 Bxd4 12.Bxd4
Nxd4=)
9...Nxe4! 10.Nd5 Qa5+ 11.c3 Nc5 12.Nxc6
dxc6 13.Nxe7+ Kh8 14.Nxc8 Raxc8 15.0-0 Rfe8=
8...a6
From my experience playing this system
over the last few years, most players, never having
seen this system before, follow a safe course of
castling kingside as White usually does in the
7.Bc4 system. However, in this particular system,
White castling kingside leads to easy equality for
Black, which I take to be a significant practical
upshot of this system. We will postpone our
discussion of White’s queenside castling until the
next chapter.
8...Qc7 This move has been championed
by Grandmasters Aronin, Pogorelov, and Balogh,
49
also with the intention of expanding on the
queenside, but they did without our core idea of
...e7-e6. It is a matter of taste which move (8...a6
or 8...Qc7) is played first; usually they will
transpose. 8...Qc7 avoids 9.Nxc6 followed by
10.e5 since the e5 square is covered, but
depending on your taste you might prefer to
welcome the uncritical 9.Nxc6. When I came up
with the idea of this system I was unaware of
8...Qc7, and for some (possibly arbitrary) reason
went with 8...a6.
9.Ndb5 Qa5 10.f3 a6 11.Nd4 Qc7
transposes to our main line in the next chapter.
9.0-0
Although Black can choose whether to
play 8...a6 or 8...Qc7, whichever he chooses on
move 8 it is important to follow it up with the
other on move 9 before embarking on 10...b7-b5.
9.h3 Qc7 As per protocol. 10.0-0
(10.Nxc6?! dxc6 11.0-0 b5 12.a3 c5
13.Nd5 Nxd5 14.Bxd5?! Rb8µ
Variations like this illustrate what I meant
in the intrduction when I claimed that it often feels
like you’re playing White when you play the
Accelerated Dragon. A couple of careless moves
by White and Black is not only equal, Black is
already better. It is uncommon for White to be
forced to play under such a slim margin of error as
he must in this line.)
10...b5
 17.Qf3

11.a3
(11.f4 b4 12.Nd5 Nxd5 13.Bxd5 e6! (Black 17...Qb8! 18.Ne2 Be5 19.Rab1 Qb4!
is also to be preferred after 13...Bb7 14.Rc1 e6 20.Nc3 c4 21.Ba2 Rfd8µ 0–1 (36)Ruiz-Panjwani,
15.Bxc6 dxc6 16.c3 e5!) 14.Bb3 (14.Nxc6 dxc6 SPICE Cup 2013).
15.Bc4 Bxb2µ) 14...Bb7„ …15.f5? gxf5 16.exf5? 10...Ne8 11.f4 d6
Qe5–+) (11...d5 12.0-0 Nc7 13.Qf3 a5 14.Bb6²)
11...Na5 12.f4 d6 The usual way to meet 12.0-0 Qc7
White’s f2-f4.
(12...Bb7!? 13.e5 Ne4 14.Nd5 (14.Nxe4
Nxb3 15.Nxb3 Bxe4³) 14...Qd8 15.Nf3 (15.Ba2
e6µ) 15...Nxb3 16.Bb6 (16.cxb3 d6µ) 16...Qb8
17.Bc7 Qe8 18.cxb3 Ng3 19.Re1 Nh5∞)
13.Qd3
(13.Nd5 Nxd5 14.exd5 Bb7µ)
13...Bb7 14.Rad1 Rac8³ ‘

13.Qe2 White’s best attempt.

a) 13.exd6 White cannot hope for an
advantage without the possession of any central
pawns 13...exd6 14.Bd4 Nf6 15.Qf3 a5„;
b) 13.Qf3 This forcing variation is worth
committing to memory, or at least remembering
that it exists and that the complications favor
Black 13...dxe5 14.fxe5 Bxe5 15.Bxf7+ Kg7–+
White has nothing to do. Black will
continue to build pressure, keeping an eye out for
central counterplay with ...e7-e5’;
9.Nxc6 bxc6 10.e5
(10.0-0 d6 11.h3 a5 12.a4?! Ba6 13.Re1
Nd7³ 14.Qd2 Nc5 15.Bxc5?! dxc5 16.Qe3 Bd4
50
 Black will simply continue ...b7-b5,
...Bc8-b7, ...Nc6-a5, and if White plays f2-f4
Black will prevent e4-e5 with ...d7-d6.
9...b5?! The first file I made on this system
recommended this move. I sent the file to a friend
over Facebook (a GM who will remain unnamed)
and he found the following slight edge for White.
In a way, the problems with 9...b5 make Black’s
life easier, because he can follow the simple rule:
first 8...a6/9...Qc7 (or the other way around) and
only then consider ...b7-b5.

Black threatens to win a piece with ...Bxc3

and ...e7-e5, as well as with ...Ne8-d6. 16.Ne2
(16.Kh1 Bxc3 17.bxc3 e5–+; 16.Rae1 Nd6
17.Bh6+ Kxh6 18.Qe3+ g5 19.Qxe5 Qa7+ 20.Kh1
Rxf7–+) 16...Nd6 17.Bh6+ Kxh6 18.Qe3+ g5
19.Qxe5 Qa7+ 20.Kh1 Nxf7–+;
13...dxe5 14.Bc5! exf4!
(14...Be6 This may be enough for equality
but is unnecessary given 14...exf4! 15.Rae1
(15.Bxe6 fxe6 16.fxe5 Rxf1+ 17.Rxf1 Qxe5=)
15...Bxb3 16.axb3 Nd6 17.fxe5 Nb7 18.Bf2 Nd8
19.Na4 Ne6∞)
15.Bxe7 10.Nxc6!
(15.Qxe7 Bd4+ 16.Kh1 Bxc5 17.Qxc5 (10.f3 Qc7 11.Qd2 (11.a4 b4 12.Na2
Be6=) (12.Nd5?! Nxd5 13.Bxd5 (13.exd5 Qe5–+)
15...Bg4 16.Qe1 Nf6 17.Bxf8 Rxf8© 13...Bb7 14.Rc1 e6! 15.Bb3 Be5! 16.f4 (16.h3
Nxd4 17.Bxd4 Bf4 18.Rb1 a5³; 16.g3 f5!µ)
16...Nxd4 17.fxe5 Nxb3 18.cxb3 Qxe5 19.Qxd7
Qxe4 20.Rf3 Qd5 21.Qe7 Rae8 22.Qxb4 Rc8=)
12...Rb8 (12...Nxd4 13.Bxd4 a5?! 14.c3² Black
should avoid positions like this where he lacks
counterplay) 13.Qd2 Nxd4 14.Bxd4 Ne8! 15.Bxg7
Nxg7

As is often the case in Dragon setups,

White’s dark squared bishop is worth the
exchange, especially here where Black is up a
pawn as well.
9...Qc7
51
 Also good for Black is 14...Rb8 15.b4?!
(15.f5 Nc4„)
15...e6! 16.bxa5 exd5 17.exd5 Bb7 18.f5
Be5 19.Qb3 Rbc8³
15.b3 e6! 16.Bxb7 Nxb7 17.b4

Black is to be preferred: the knight will

find its way to c5 via e6, and White’s pieces are
sloppy on b3 and a2.) 11...Na5 12.Nd5 Nxd5
13.Bxd5 Rb8=)
10...dxc6 11.Qxd8
(11.f3 Qc7 12.a4 (12.Qe1 c5 13.a4 b4
14.Ne2 c4 15.Ba2 a5³) 12...Rd8 13.Qe1 b4 14.Ne2
a5 15.Qf2 (15.Bc4 Nd7³) 15...Nd7=) 17...e5!
11...Rxd8 12.Rfd1 Nd7 13.f4 White has a
moderately annoying clamp on Black’s position. 17...Rad8?! 18.f5 exf5 19.exf5 Rfe8 20.f6
Black should still be fine with precise play but it is Bf8 21.Bf4 Qb6 22.a4 Nc5 23.a5 Nxd3 24.axb6
better to avoid this. Nxf4 25.Rxf4± Kurnosov,I-Topalov,V Astana
10.f4 d6 11.h3 KAZ 2012.
18.Ne2

18.fxe5 dxe5³
18...d5! 19.fxe5 Qxe5 20.Bd4

20.exd5 Rae8–+
20...Qxe4µ

11...Na5 12.Qd3 b5 13.Nd5

13.a3 Bb7 14.Rad1 Rac8³ White has

nothing to do. Black will continue to build
pressure, keeping an eye out for central
counterplay with ...e7-e5.
13...Nxd5 14.Bxd5 Bb7
52
 2

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6

5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.f3

White signals his intention to play in

Yugoslav Attack fashion with Qd2 and 0-0-0. This
is the most critical continuation, but since White
players are not prepared for this new system, over
the board they tend to shy away from the
challenge.
9...Qc7

Now White’s b3 Bishop literally has no

squares to go to, a triumph of our concept 13.a3
Bb7„ Superficial developing moves on White’s
part have allowed Black to take over the initiative:
...d7-d5 is now imminent.;
10.g4!? Black must be accurate here.

9...b5?! Don’t forget — first 9...Qc7, and

only then 10...b5. 10.Nxc6 dxc6 11.e5²
10.Qd2

10.0-0 It makes no sense to combine f3

with kingside castling, as White’s only
constructive plan after 10.0-0 is to go for f4 10...b5!
anyway. (10...e6 Black can play this way and
10...Na5 transpose to the main line but 10...b5! takes
(As usual 10...b5!? is perfectly playable as advantage of White’s premature 10.g4. 11.Qd2 b5
well.) 12.0-0-0
11.Qd2
(11.Nd5?! Nxd5 12.exd5 b5 (12...Nxb3
13.axb3 Qe5 14.Re1 Qxd5 15.c4©) 13.Qd2 Bb7³)
11...b5 12.Rad1 e6!

53
 12...Na5?! (¹12...Bb7 13.h4 h5 This
transposes to our main line.) 13.h4‚ 13...h5 We
will see similar positions later on, except with
Black’s knight on c6.
a) 13...Bb7?! 14.h5 In this system it is
almost always unwise for Black to allow White
h4-h5. 14...Rac8 (14...b4 15.Na4 Nxe4 16.fxe4
Bxe4 17.hxg6 Nxb3+ 18.axb3 fxg6 19.Qh2±)
15.Kb1±
15...Nxg4 16.fxg4 Nxb3 17.Nxb3+–;
b) 13...d5!? It is worth researching this
move further;
14.Bh6±
54
It took me a while to realize that Black
should keep his knight on c6 to prevent White
from playing this move, on account of ...Nxd4
winning a piece. Allowing White Be3-h6 in this
way is enough of a concession to turn a
dynamically equal position into a difficult one for
Black. (14.g5 Ne8„))
11.g5
11...Nh5
(11...Ne8 Playable but not the best. 12.Nd5
Qd6 (12...Qe5 13.c3± Na5?? 14.Bf4+– This is
why the knight is better on h5) 13.Qd2²)
12.Nd5 Qe5 13.c3 Na5! 14.Ne2 (14.Bc2
Nc4µ) 14...Nxb3 15.axb3 (15.Qxb3 e6 16.Ne7+
Kh8³) 15...Bb7„
10...b5
I give some analysis on 10...Na5 in the
next chapter, but that material is an appendix to
this one; I prefer to play in the way I recommend
here.
 11.0-0-0
11.Nxc6 dxc6= Although objectively
equal, Black is to be preferred in practice. White is
planless; the b3 bishop is vulnerable to Black’s
...c5-c4, and the c3 knight no longer has its
favorite d5 square.;
11.a4 b4 12.Na2 Rb8! 13.Nxc6
(13.0-0 Nxd4 14.Bxd4
14...Ne8! 15.Bxg7 (15.Nxb4 Bxd4+
16.Qxd4 e5 17.Qc3 Qb6+–+) 15...Nxg7= The
knight will find its way to c5 via e6; White’s
pieces are sloppy on b3 and a2.)
13...dxc6 14.Nxb4 c5 15.Nd5
(15.Nd3
15...Nxe4!! (15...c4 16.Bf4±) 16.fxe4 c4
17.Bf4 e5³ The ability to play ...e5 in response to
Bf4 is why Black must sacrifice with 15...Nxe4!!)
15...Nxd5 16.Qxd5 Bxb2 17.Rd1
(17.Rb1?! Ba3 18.0-0! Be6µ)
17...Be6 18.Qxc5 Qxc5 19.Bxc5 Bf6
55
20.Bxe6 fxe6= …...Rfc8;
11.g4 Bb7 12.g5 Nh5 13.Nd5
13...Qd6 It is admittedly a little difficult to
remember to play 13...Qd6 in this 11.g4 variation,
whereas in the 10.g4 variation the right move was
12...Qe5. The reason is that in the 10.g4 variation
White was forced to play 13.c3 in response to
12...Qe5, whereas here, thanks to White’s
d2-queen already being developed, White can
defend the d4-knight with 14.0-0-0.
(13...Qe5 14.0-0-0²)
14.0-0-0 Nxd4 15.Bxd4 Bxd4 16.Qxd4
Bxd5 17.e5 Qc7 18.Qxd5 Ng7=;
11.h4 h5 12.0-0-0 Bb7 transposes to the
main line.
11...Bb7
Although this position has only been
reached 30 times in the datababse, it is the first
critical position of our main line. Both sides have
developed their pieces and connected their rooks,
and it remains to be determined which of the
opposing wing attacks will succeed. Black’s sights
aren’t limited to the queenside though, because the
central pawn majority offers prospects for
counterplay with ...d7-d5 or even ...e7-e5.
12.h4

By far the most common move.

12.Nd5?! Nxd5 13.exd5 Ne5 14.Kb1 Rac8
15.Bh6 Nc4 16.Bxc4 Bxh6 17.Qxh6 Qxc4 18.h4
f6µ;
12.Bh6?? Nxd4 13.Bxg7 Nxb3+ 14.axb3
Kxg7–+;
12.Kb1
No longer check as it was when the White
king was on c1. 16.Qh6+–)
13.Bh6
(13.h4 b4!

This is the concession we induce by

prolonging the Knight’s stay on c6. White now
threatens 13.h4 h5 14.Bh6, as after 14...Nxd4
15.Bxg7 Nxb3 is no longer check, allowing
16.Qh6+–.
12...Na5! Having induced White’s Kb1,
Black can now go for lines similar to those after
10...Na5 where White is effectively down a tempo
in the critical variations because Kb1 turns out to
be unnecessary.
(12...e6 13.Nxc6 Bxc6 14.Bf4 Qb7 15.Nd5
a5 16.Ne7+ Kh8 17.a3²; 12...Rac8?! 13.h4 h5
14.Bh6 Nxd4 15.Bxg7 Nxb3??
Since White spent a tempo on Kb1,
Black’s queenside counterplay is in time. 14.Na4
(14.Nd5 Bxd5 15.exd5 Rfc8„ 16.h5?! Nxb3
17.Nxb3 Nxd5!µ) 14...e5! 15.Ne2 d5!„)
13...Nxb3 14.cxb3
(14.Nxb3 Bxh6 15.Qxh6 b4 Black can
allow White’s Nc3-d5 when he can capture with
the bishop on b7, leaving the f6 knight to defend
against White’s h4-h5 attack (in particular the h7
square will be defended). 16.Nd5 Bxd5 17.exd5
Rfc8 18.Rd2 Qd6 19.h4 a5 20.h5 a4 21.Nd4 a3
22.hxg6 (22.b3 Ra5 23.hxg6 fxg6–+) 22...fxg6–+)
14...b4 15.Bxg7
(15.Nce2 Bxh6 16.Qxh6 d5!

56
 17.h4 e5 18.Nf5 Fear not! 18...Ne8!
19.Ne3 (19.Qg5 f6 20.Nh6+ Kg7 21.Nf5+ Kh8³)
19...d4 20.Nc4 f6=)
15...bxc3 16.Qh6 Rfc8!
This is one of the few lines worth
committing to memory as well. 17.bxc3 Nh5
18.Nf5 gxf5 19.Bd4 e5 20.Qxh5
(20.Bb6 Qxc3 21.Qxh5 Qc2+=)
20...exd4 21.Qg5+
(21.Rxd4 Qxc3 22.Qg5+ Kh8 23.Qf6+
Kg8 24.Qg5+ Kh8=)
21...Kh8 22.Qf6+ Kg8=;
12.g4
57
A slight inaccuracy. White should start
with 12.h4, so as to ensure that Black plays
...h7-h5; in this line Black can get away without it.
12...Na5!
(12...Rac8 13.g5 Nh5 14.Nd5²; 12...e6!?
Black can transpose to the main line with this
move. 13.h4 h5 etc.)
13.Kb1
a) 13.Bh6 Bxh6 14.Qxh6 Rac8„ Black’s
queenside attack is too fast for White (14...b4
15.Nd5 Nxb3+ 16.axb3 Nxd5 17.exd5 Rac8
18.d6!∞);
b) 13.g5 Nh5³ Black has a knight on each
rim, but it is White’s position which is dim.;
c) 13.h4 Rfc8!
We see that White’s g2-g4 is redundant;
Black could have played h2-h4-h5 without it, as it
would have been suicide for Black to capture
...Nf6xh5. (13...h5?! 14.Bh6²) 14.h5 Nxb3+
15.axb3 b4 16.Na4 e5 17.hxg6 fxg6 18.Ne2 Nxe4!
19.fxe4 Bxe4„;
 13...Rac8
(13...Rfc8 14.Bh6 Bxh6 15.Qxh6 b4
16.Nf5 gxf5 17.g5±)
14.Bh6
As a rule, Black’s counterplay (whether on
the queenside or in the center) tends to ‘work’
when White has spent a tempo on Kb1. 14...Nxb3
(14...Bxh6 15.Qxh6 Nxb3 (15...b4 16.Nd5
Bxd5 17.exd5 (17.Bxd5 e6 18.Bb3 d5 19.exd5
Nxb3 20.d6 Qc5³) 17...Nxb3 18.d6!! Black may be
fine here but to allow this is unnecessary 18...exd6
19.cxb3± White threatens Nd4-f5, winning.)
16.axb3 transposes to 14...Nxb3. (16.cxb3? b4
17.Nce2 Nxg4! 18.fxg4 Bxe4+ 19.Ka1 Bxh1
20.Rxh1 Qe5µ))
15.axb3
(15.cxb3 b4 16.Nce2 Nxe4 17.fxe4 Bxh6
18.Qxh6 Bxe4+ 19.Ka1 Bxh1 20.Rxh1 Qe5µ
21.Rd1 (21.h4 Rc5 22.h5 Qg5–+) 21...Rc5–+)
15...Bxh6 16.Qxh6 b4 17.Rhg1!?
(17.Na4 d5!„)
58
17...Kh8! 18.Na4
(18.Na2 d5 19.exd5 Nxd5µ)
18...d5 19.e5 Qxe5 20.g5 Nd7µ
12...h5!N
A new move, but more importantly a new
concept for this line: containment.
Previously, Black would rush with
queenside counterplay, hoping to mate White
before getting mated himself. It turns out that
Black can contain White’s kingside attack and
solidify the center with ...e7-e6, and only then
proceed with queenside and/or central counterplay.
For his part, White must keep the attack going
with urgency, because Black’s queenside
counterplay is just a few moves away, and it will
come with devastating effect (...Rac8, ...Rfd8,
...Na5, etc.).
12...Na5!?
This may turn out to be playable as well,
but it is not in the spirit of the concept I am
advocating. To allow White h4-h5 without
immediate counterplay is akin to the ordinary
Yugoslav Attack in the Dragon; our aim is to play
in ‘Taimanov style’ with ...h7-h5 and ...e7-e6.
13.h5 b4 14.Na4
(14.Nce2 Rfc8 15.hxg6 Nxb3+ 16.Nxb3
hxg6 17.Nc5 Bc6 18.Bh6 Bh8∞)
14...Nxb3+ 15.Nxb3 Nxh5 16.Nb6
(16.g4 Nf6 (16...Ng3? 17.Qh2) 17.Nb6
Rad8 18.Na5∞)
16...Rad8 17.Na5²
13.g4
The character of the position demands that
this be played. Timidity will not suffice; time is of
the essence in the race to determine which of the
flank attacks will succeed. My initial attitude
towards this system was that I should delay
...e7-e6 for as long as possible, because I felt I
would always have the option later on, and I
thought there could be more useful attacking
moves on the queenside such as 13...Na5 (which
also guards the important d5-square). However, as
I delved deeper into this variation, I realized that
Black should keep the knight on c6 for a little
longer to deter White’s Be3-h6. The result is a
rather peculiar middlegame ‘standoff’, where
White cannot make progress with Be3-h6 until
Black moves his knight from c6, but at the same
time Black cannot make progress on the queenside
until he does either.
13.Nxc6 This may look unnatural — and it
is — but the ‘computer move’ which it is based on
(14.Bd4!) is not to be underestimated. Black’s
f6-knight is the MVP (most valuable piece) of his
59
position, as without it, White’s g2-g4xh5 would be
decisive. Fortunately, the f6-knight is difficult to
get at, as 13.Bg5?? loses to 13...Nxe4. White’s
13.Nxc6 takes aim at Black’s f6-knight by paving
the way for 14.Bd4. 13...Bxc6
(13...dxc6 Black does not quite equalize
here. 14.Bc5 a5 15.a3 (15.a4 Ba6∞) 15...Rfd8
(15...b4 16.Na4 Ba6∞) 16.Qe3 Rxd1+ 17.Rxd1 a4
18.Ba2 Rd8 19.Rxd8+ Qxd8 20.e5 Nd5 21.Nxd5
cxd5 22.f4²)
14.Bd4
(14.Bh6 a5 15.Bxg7 Kxg7 16.Qd4 b4
17.Na4 d6³ 18.g4?! hxg4 19.h5 Qa7! 20.Qd2
Rh8µ)
14...a5!
(14...e6?! 15.e5 (15.g4 a5 16.gxh5 Nxh5=)
15...Nh7 16.Qe3 a5 17.a4 Rfb8 18.axb5 Bxb5
19.g4 White’s attack is too fast. 19...a4 20.Ba2
hxg4 21.h5 a3 22.b3 Bc6 23.hxg6 fxg6 24.Ne4
Rf8 25.f4‚)
15.e5
(15.Nd5?! Bxd5 16.Bxd5 Rac8µ)
15...Ne8 16.Nd5
(16.Qg5 a4 17.Nd5 Black has an
embarrassment of riches, needing to choose
between two beautiful variations 17...axb3!!
(17...Bxd5 18.Bxd5 Rc8 19.Rd2 e6 20.Be4 Qc4
21.Kb1 Nd6!!„) 18.Nxc7 bxa2 19.Kd2 Nxc7³)
16...Bxd5 17.Bxd5 Rc8

Fair warning is in order: what follows is a
total mess. That being said, look closely and you
will see that it is White who needs to be accurate
and find several (nontrivial) ‘only’-moves in order
to survive the complications. As Black, playing a
risky line like this, we should welcome our
opponents into what Tal called the “deep dark
forest where 2+2=5 and the path leading out is
only wide enough for one”. 18.f4
(18.g4 e6 19.Be4 Bxe5 20.Bxe5 Qxe5
21.Rhe1 Qc5 22.Qh6 Nf6 23.Bxg6 fxg6 24.Qxg6+
Kh8 25.g5 Nd5 26.Qh6+ Kg8 27.Qg6+ Kh8=)
18...d6 19.Qe2
(19.e6? f5³)
19...dxe5
(19...e6 20.Bf3 Bh6 21.g3 d5 22.g4 Bxf4+
23.Kb1‚)
20.fxe5 Qb8!!
Making room to relocate the knight from
e8-c7-e6. 21.g4
(21.a3 Nc7 22.Ba2 b4„)
60
21...Nc7 22.Bf3 Ne6 23.gxh5 Rc4 24.Bc3
(24.c3 Rfc8 25.Kb1 b4 26.Bd5 bxc3
27.Bxc4 Rxc4 28.Bxc3 Rxc3 29.hxg6 fxg6³)
24...Rfc8!
(24...Qc7 25.hxg6 (25.Kb1 Rxc3 26.bxc3
Nc5 27.hxg6 Na4 28.gxf7+ Rxf7 29.Qe3 Nxc3+
30.Kc1 Nxa2+ 31.Kd2 Bxe5∞) 25...Rxc3
26.gxf7+ Kh8 27.bxc3 Qxc3∞)
White’s only move to not be clearly worse
here is... 25.Rd7™ 25...Qb6 26.hxg6 Nf4 27.gxf7+
Kf8 28.Qd1™ 28...b4 29.Kb1™ 29...bxc3 30.b3
Qc5 31.Rd8+
(31.Ka1 Qa3 32.Rd8+ Rxd8 33.Qxd8+
Kxf7 34.Bd5+ e6 35.Qd7+ Kg8 36.Bxe6+ Nxe6
37.Qxe6+ Kh7 38.Qf5+ Kg8 39.Qe6+ Kh7=)
31...Rxd8 32.Qxd8+ Kxf7 33.Rf1™
33...Rd4 34.Bh5+ Ke6 35.Qg8+ Kd7 36.Bg4+ Kc7
37.Qc8+ Kb6 38.Qb8+ Kc6 39.Bf3+
(39.Qc8+ Kb6=)
39...Kd7 40.Re1 Qa3 41.e6+ Nxe6
42.Qb5+ Kd8 43.Qb8+ Kd7=;
13.Rdg1?! White cannot hope for an
advantage with such timidity 13...e6 14.g4

14...Qd6! 15.Nxc6 Qxd2+ 16.Bxd2 dxc6
17.gxh5 Nxh5³ White’s Bb3 is a mere spectator,
soon to be cornered by Black’s c-pawn (...c5-c4).
13...e6!
Like 13...Na5, this move also prevents
White’s g4-g5 followed by Nc3-d5, but by
keeping the knight on c6, White’s Be3-h6 is
hindered. What follows is somewhat a game of
‘cat and mouse’ — White’s main resource of
Be3-h6 is unavailable at present, and it is difficult
to see how to continue the attack without that.
Black would welcome White’s g4-g5, which locks
up the kingside and directs Black’s knight to d6
via e8. On the other hand, Black’s queenside
attack is stalled by the fact that he can’t play
...Nc6-a5 without allowing Be3-h6. That being
said, whereas White is at a loss for ways to
improve his position, Black can place his rooks on
the d-and-c (or b-) files and threaten both central
counterplay and a queenside pawn storm.
61
13...Na5?!
14.Bh6! This is ultimately the problem
with 13...Na5 — White gets to exchange Black’s
‘Dragon bishop’. White threatens Nd4-f5, so the
following (14...e6) is forced.
a) 14.gxh5 Nxh5 15.Rhg1
Computers take some time to appreciate
the power of this idea, but anyone who has studied
the Yugoslav Attack will know that White’s recipe
is to, as Fischer said, “pry open” files to Black’s
king, then “sac, sac, mate”. 15...e6! (15...Nxb3+
16.axb3 b4 17.Na4 d5 18.Rg5! This is the point.
White really does intend to “sac, sac, mate”
18...dxe4 19.Rxh5! e5 (19...gxh5 20.Rg1+–)
20.Qh2 gxh5 21.Nf5+– White’s attack is
overwhelming) 16.Rg5!
 The aggressively inclined Black player
may be in dismay — White is initiating dangerous
threats like Rg5xh5, and Black’s counterattack
seems far off. However, as Suba teaches in his
book Dynamic Chess Strategy, “initiative is a
psychological advantage”, so it is up to us as
defenders to grant or deny this advantage to our
opponents. The aggressor is not better simply by
virtue of being on the offensive: White was the
first to initiate action but if we defend well, he
may quickly run out of steam and be left with no
attack and static pawn weaknesses. 16...Kh7!
(16...Bf6?! 17.Rxh5 gxh5 18.f4±) 17.Rdg1 Bf6
18.Rxh5+ gxh5 19.e5 Nxb3+ 20.axb3 Bxe5
21.Rg5 Rg8
This position is a draw according to my
computer. Technically, then, Black is not worse in
this variation, but you can see why I prefer to
avoid this mess by playing 13...e6!. 22.Rxh5+
(22.Nxe6 dxe6 23.Rxh5+ Kg6 24.Rg5+ Kh7=
(24...Kf6 25.Rxe5 Qxe5 26.Bd4
62
26...Rg1+ 27.Nd1 Bxf3 28.Bxe5+ Ke7!!
Study-like 29.Qb4+ Ke8 30.Kd2 Rd8+ 31.Ke3
Bxd1=) ) 22...Kg6 23.Qd3+ Kf6 24.f4 Bd6∞ The
Higher Power calls this “0.00”, but I would not
write a book recommending this madness as the
first option.;
b) 14.g5 Black is usually happy to see this;
White blocks off his own attack and lures Black’s
knight to e8 where it wants to go anyway
14...Ne8! 15.Nde2 Nd6!
When the concept of transporting this
maneuver from the Taimanov occurred to me, I
was pessimistic that it would turn out to be
possible in the Accelerated Dragon. It was a
pleasant surprise when I saw that the computers
approve of it, and as it turns out it is a recurring
theme in this line. 16.Bf4 b4 17.Na4 (17.Nd5?!
Nxb3+ 18.axb3 Bxd5 19.Qxd5 Rfc8 20.Rd2 a5!
21.Bxd6 exd6 22.Qxd6 a4 23.Qxc7 Rxc7 24.Kb1
a3!µ) 17...Rfc8 18.Bxd6 exd6 19.Kb1
 19...d5! 20.exd5 Nxb3 21.cxb3 Bxd5 Not an easy move for humans to find
22.Qxd5 Qc2+ 23.Ka1 Qxe2„; because we tend to expect that the battle will be
14...e6 decided on the kingside. The logic of this move is
(14...Nxb3+ clear: White threatens e4-e5, and Black cannot
prevent this with 15...d6 for tactical reasons.
Furthermore, White’s pressure on the kingside is
still felt, and eventually some sort of Nd4-f5 or
gxh5 will be decisive.
(15.Bxg7 This eases Black’s task a little;
there is no need to take on g7, as Black certainly
couldn’t allow White’s queen to infiltrate by
taking on h6 15...Kxg7 16.g5 (16.Rhe1 Nxb3+
17.axb3 d6²) 16...Ne8! (16...Ng8!? 17.Rhe1 b4
(17...Ne7 18.f4 Rad8 19.f5 Kg8 Black’s position is
surprisingly resilient) 18.Na4 e5 19.Ne2 Nxb3+
20.axb3 Rac8 21.Kb1 a5 22.c4 Bc6=) 17.Kb1

In the one tournament game in which I had

this position over the board, I played this move. I
took over the advantage in that game, but only
because my opponent let me...15.axb3 b4 16.Nce2
hxg4 17.Bxg7 Kxg7 18.Qg5 Nh7 19.Qxe7?!
(19.Qxg4 e6 20.Nf4±) 19...Qa5 20.Qxd7? gxf3!
21.Nf4?! Qa1+ 22.Kd2 Qxb2 23.Nde6+ Kg8
24.Nxf8 Qc3+ 25.Kc1 Bxe4µ Cao,J-Panjwani,R
Kitchener 2015).
15.Rhe1!

17...Rd8! Preparing the thematic ...Ne8-d6,

since now after 18.Nxe6+ dxe6 the d6-Knight is
defended 18.Nde2 Kg8! Threatening ...b5-b4
(18...b4 19.Qd4++–) 19.Nf4 b4 20.Na4 Nxb3
21.cxb3 d5! 22.exd5 a5!„)
63
 15...Nxb3+
(15...b4 16.Na4 hxg4 17.Bxg7 Kxg7 18.e5
Nd5 19.fxg4±; 15...d6?? 16.Bxe6! Bxh6 17.Qxh6
fxe6 18.Qxg6+ Qg7 19.Qxg7+ Kxg7 20.Nxe6++–)
16.axb3 d6 17.Bxg7 Kxg7 18.gxh5 Nxh5
19.Rg1

14.a3 This is admittedly a ‘computer

move’ — I can’t imagine many humans will
weaken their king position unprovoked.
Nevertheless, it is instructive to see how Black
builds up attacking potential. 14...Rfc8!

Black is suffering. White’s attack is not

decisive, but lacking counterplay Black is
condemned to passive defense. 19...Kh7
(19...Rae8 20.b4 Black needs to be
accurate to not collapse here. 20...Nf6?! 21.h5!
Nxh5 22.f4 d5 23.Rdf1 dxe4 24.f5!±; 19...Rh8??
20.Rxg6++–)
20.Qg5 Threatening Nd4xe6, as well as
f4-f5.
(20.Rg5 Rg8 21.b4 Rad8 22.Nde2 f6
23.Rg4 e5!=)
20...Rfe8 21.b4! Rac8
(21...Qe7 22.Nb3±)
22.Kb1 Qe7 23.Nb3± Black is positionally In such positions it is notoriously difficult
dominated. to decide which rook to put on a given file. Pal
14.Kb1 Benko is rumored to have joked that his method is
as follows: think really hard, decide which rook
Small ‘refinement’ moves like this are would be the right choice, then play the other
useful prophylaxis for when the position rook! By placing the f-Rook on c8 instead of d8,
(inevitably) explodes. Black sidesteps a potential Be3-g5 pin, and also
defends his Qc7 in case, after ...Nc6-e5 for
example, White pins the Knight with Be3-f4 or
Qd2-h2.
(14...Na5?! This always runs into Be3-h6
15.Bh6± (15.g5 I can’t resist including some pretty
variations which I encountered in my research.
15...Ne8 16.Ba2

64
 16...Nd6! Computers rarely suggest this as
their first choice (including here) but from my
perspective, if I can get away with this maneuver
then I’m playing it 17.Qh2 (17.Nxe6 fxe6 18.Qxd6
Qxd6 19.Rxd6 Bxc3 20.bxc3 Rxf3 21.Bb6 Nc4
22.Bxc4 bxc4 23.Rxd7 Bxe4=) 17...e5! 18.Nb3
Nac4 19.Bf2 a5
Black’s pieces are the picture of harmony;
in contrast, look at White’s! 20.Nd5 Bxd5
21.Rxd5 To those who complain that computers
have ruined chess, I sympathize for the most part,
but every now and then they win back our hearts
with ingenious sequences like the following
21...Nxb2!! 22.Kxb2 Nc4+ 23.Ka1 d6!! White’s
pieces are so uncoordinated that Black can take
time out of his day for this calm, solidifying move,
covering the c5 square and threatening ...Nc4xa3.
(23...Nxa3 24.Bc5±) 24.c3 b4! 25.a4 Qc6 26.Rb1
Qxa4 27.Be1 Rfb8
65
Alas, at least as often as computers
uncover hidden beauties for us, they ruin our fun
with flawless defense, insisting that chess played
correctly settles down to a draw. 28.Qe2 Na3
29.Rc1 Nb5 30.Rd3 Bf8= Apparently 0.00 but we
needn’t continue along these lines, as 15.Bh6 must
not be permitted earlier on.))
15.Bf4
a) 15.Nxc6 Bxc6 16.Bf4 Qb7 17.Be5 b4!
18.axb4
18...a5!! 19.Qf4 (19.bxa5?? Rxa5 20.Qf4
Rxe5! 21.Qxe5 Nxe4–+) 19...axb4 20.Bxf6
(20.Nb1 Nxe4!! 21.Bxg7 (21.fxe4 Bxe5 22.Qxe5
Bxe4 23.Rh2 hxg4©) 21...Nc5!! 22.Bc4 Kxg7
23.gxh5 Bxf3 24.Rdf1 Bxh5 25.Qxf7+ Kh8
26.Qf6+ Kh7=) 20...bxc3 21.bxc3 Ra5!³ Black
enjoys superior king safety and White’s
light-squared bishop is caught offside, unable to
assist on the Kingside;
b) 15.Kb1 Ne5 16.Bg5 hxg4 17.Bxf6 Bxf6
18.h5 g5 19.fxg4 Qc5! 20.Rdf1 Qe7„
 Black plans to play ...Rc8-c5 and double
up rooks on the c-file, threatening to sacrifice the
exchange on c3 with devastating effect.;
15...Ne5 16.Qh2 d6
Black has a lot going for him: more central
pawns, White’s attack has been halted, and
Black’s queenside attack is just getting started. In
addition to this, the tactics on e6 do not work for
White.
17.Nxe6
(17.Bxe5 dxe5 18.Nxe6 fxe6 19.Bxe6+
Kh7 20.Bxc8 Rxc8„)
17...fxe6 18.Bxe5 dxe5 19.g5 Qe7 20.gxf6
Bxf6 21.Qh3 Rc6 22.Qg3 Kh7=;
14.g5 Ne8
66
15.Nde2
a) 15.a3 Na5 16.Kb1 (16.Ba2 Nc4 17.Bxc4
bxc4=) 16...d5! 17.exd5 Nxb3 18.cxb3
18...Rd8! We have seen this before; it is
worth keeping this idea in mind. 19.Rc1 (19.dxe6
Bxf3µ) 19...Qb8=;
b) 15.Nxc6 Bxc6 16.Bc5 d6 17.Bxd6?
(17.Bd4 Bxd4 18.Qxd4 Rb8„) 17...Nxd6 18.Qxd6
Qb6!µ with ...a5-a4 to follow;
15...d6 16.Rhf1 Na5 17.Kb1
(17.f4? b4 18.Na4 Nxb3+ 19.axb3 Bxe4µ)
17...Rd8! Since White’s pawn storm has
been blockaded, he is condemned to meandering
piece play, whereas Black is effectively up a pawn
in the center. 18.Bd4 d5 19.exd5 Nxb3 20.axb3
Bxd5=;
14.Nxc6 Bxc6 15.Bf4 Qb7
 they used to think that the more talented player has
a moral responsibility to initiate aggression!
15.Bh6 White invites simplifications which
yield no advantage 15...Bxh6 16.Qxh6 Nxd4
17.Rxd4 Qe5!

16.Bh6 a5 17.Bxg7 Kxg7 18.g5

(18.gxh5 Nxh5 19.Rhg1 a4 20.Bd5 b4
21.Qd4+ Kh7 22.Bxc6 dxc6 23.Ne2 c5!„)
18...Ne8 Computers take quite some time
to realize that black has no problems here. 19.Bd5
b4 20.Bxc6 Qxc6 21.Ne2 Rd8=; 18.gxh5
14.Rdg1 Qd6! 15.Nxc6 Qxd2+ 16.Bxd2 (18.Qd2 hxg4 19.fxg4 b4 20.Rxb4 Nxe4
Bxc6 17.g5 Ne8„ 21.Nxe4 Bxe4 22.Rf1 d5 23.Rb7 Rf8=
14...Rfd8

14...Rad8 The following is a good

illustration of what not to do. 15.a3 d5?! 16.gxh5
Nxh5 17.Nxc6 Bxc6 18.exd5 exd5 19.Bg5 Rd7
20.Ne2 Bb7 21.Nd4±;
14...Rfc8?! This doesn’t work here, as
White hasn’t weakened himself with a2-a3, so
Black’s queenside pawn storm is less effective
than in was after 14.a3. 15.Nxc6
(15.Bf4 Ne5 16.Qh2 d6 17.Nxe6 fxe6
18.Bxe5 dxe5 19.Bxe6+ Kh7 20.g5 Ne8 21.Bxc8
Rxc8=)
15...Bxc6 16.Bf4 Qb7 17.Be5 a5
(17...b4 18.Qf4 bxc3 19.Bxf6 Qc7 The computer gives ‘zeroes’ (“0.00”) but I
20.Qg5±) would prefer Black in a practical setting because
18.Qf4 a4 19.Bxe6 dxe6 20.Bxf6 b4 Black has a better Bishop, more central pawns,
21.Bxg7 Kxg7 22.Ne2± and potential for taking over the initiative with
15.g5 ...a5-a4, whereas attempts by White to ‘do
something’ seem to only make his position worse.
We are in unexplored territory here, so I 24.g5?! a5 25.a4 Rab8 26.Rb5 Rxb5 27.axb5 Qc7³
can only guess what White’s likely choice would With ...Rf8-b8 to follow.)
be in an over-the-board situation. I suspect that the 18...Nxh5 19.Rhd1 Qf4!
urge to ‘do something’ would be overwhelming, (19...Bc6 Black should be able to hold here
and White would be the first to cave in. These with accurate play but the position is a little
days we tend to think the onus is on White to annoying to defend because Black lacks a
initiate complications. In the early days of chess, constructive plan. This is why I recommend
67
forcing matters by liquidating with 19...Qf4. (26.Bc4 Ng2 27.Bxa6 Nxh4 28.Bb7 Rd7
20.Ne2 a5 21.a3 b4 (21...Re8 22.Rg1 Rab8 23.f4 29.Rxd7 Bxd7 30.Ka2 g5 31.Ne4 Kg7 32.Nxg5
Qg7 24.Qg5 d5 25.e5²) 22.a4 Re8∞) Ng6 33.Ka3 Nxe5 34.Kb4 f5 35.Kc5 Kf6 36.f4
20.Qxf4 Ng6 37.Nh3 e5=; 26.Nxa4 Ng2 27.Nc5 Nxh4
(20.Qg5 Bc6 21.Ne2 Qxg5 22.hxg5 a5 28.f4 Ng2 29.Nxa6 Nxf4=)
23.a3 Kf8=) 26...Bxa4 27.Nxa4 Rd1+
20...Nxf4 21.Rxd7 Rxd7 22.Rxd7 Bc6= (27...Ng2 28.Rc4 a5 (28...Rd5 29.Rc8+
Kg7 30.c4 Rxe5 31.c5±) 29.c3 Rd3 (29...Kg7
30.b4 axb4 31.cxb4 Ne3 32.Rc3 Nd5 33.Rb3²)
30.b4 axb4 31.cxb4 Rxf3 32.Kb2 Ne3 33.Re4 Nd5
34.b5 f5 35.exf6 Rf5 36.Ka3 (36.Rxe6 Nc7=)
36...Nxf6 37.Rxe6 Kf7 38.Rb6 Nd7 39.Rb7 Ke6
40.Nc3 Rh5 41.Rc7 Rxh4 42.b6 Nxb6 43.Rc6+
Kf5 44.Rxb6=)
28.Ka2 Rh1 29.Nc5
(29.b4 Rxh4 30.Ra7 Nd5 31.c3 Nxb4+
32.cxb4 Rxb4 33.Nc5 Rf4 34.Rxa6 Rxf3=)
29...Rxh4 30.Nxa6 Nd5 31.Rc8+ Kg7
32.Nc7 Ne7 33.Ne8+ Kf8 34.Ra8 Rf4 35.Nf6+
Kg7 36.Ra3 Nc6 37.Ng4 g5=;
15.Bf4 e5
It is not necessary to memorize the precise
way Black draws in the ensuing variations; I
include them just for the sake of completeness, but
without comments. It is enough to recognize that
White’s h4 pawn is bound to eventually capitulate
to Black’s ...Kg8-g7 and ...Ra8-h8 or ...Nf4-g2,
and White’s Bishop remains frozen, missing in
action on b3.
23.Rc7
(23.Rd6 Rc8=; 23.Rd2 Kf8! 24.Ne2 Nxe2
25.Rxe2 Ke7 26.Rg2 Rh8 27.Rg4 e5=)
23...Be8 24.e5
a) 24.Rc5 b4 25.Nd1 (25.Na4 Bxa4
26.Bxa4 Kg7=) 25...Bb5 26.c4 bxc3 27.Nxc3
Bd3+ 28.Kc1 (28.Bc2 Kg7 29.Bxd3 Nxd3 30.Ra5 16.g5
Rh8 31.Rxa6 Rxh4=) 28...Rd8=; (16.Nf5 gxf5 17.Bg5 Na5 18.Bxf6 Bxf6
b) 24.Kc1 Rd8=; 19.Bd5 f4 20.g5 Bg7 21.g6 Kf8 22.Bxf7 Rac8µ)
c) 24.a4 Ng2 25.axb5 axb5=; 16...Nxe4!
24...Rd8 25.a4 (16...Nxd4 17.Qxd4 d6 18.Qd3 Nxe4
(25.a3 a5 26.a4 bxa4 27.Nxa4 Ng2=) 19.Nxe4 exf4 20.Nf6+ Kh8 21.Rhe1∞)
25...bxa4 17.Nxe4 exf4„;
(25...Ng2?! It is inspiring that despite such 15.Rdg1 b4 16.Na4 d5!
an error Black can still hold; such margin for error
is important to account for when choosing a
repertoire. 26.a5! Nxh4 27.f4 Ng2 28.Ra7 Nxf4
29.Ne4 Bc6 30.Nd6 Bd7 31.Rxa6 g5 32.Ne4 Ng6
33.Nxg5 Nxe5=)
26.Bxa4
68
 17.gxh5
(17.g5 Nxe4! 18.fxe4 dxe4 19.c3 bxc3
20.bxc3 Na5©)
17...Nxh5 18.Nxe6
(18.Rxg6 fxg6 19.Nxe6 Qe5 20.Nxd8
Rxd8 21.Bxd5+ Kh8=; 18.Nc5 Nxd4 19.Bxd4
Bxd4 20.Qxd4 dxe4 21.Qxb4 Bd5 22.Nxe4 Bxe4
23.fxe4 Ng3 24.Rh3 Ne2 25.Rf1 Ng3 26.Rg1=)
18...fxe6 19.Rxg6 Qf7 20.Qg2
(20.Rg5 Ne5„; 20.Rxe6 Na5„)
20...Ne5 21.Rh6 Nf4
22.Bxf4 Qxf4 23.Rxe6 Qxf3 24.Qg1
(24.Rg1 Qxg2 25.Rxg2 Kf8 26.Rf2+ Kg8
27.Rg2 Kf8=)
24...Re8 25.Nc5
(25.Rb6 Ra7 26.h5 Kh8 27.h6 Bf8µ)
25...Bc6=
15...Ne8 16.Qf2 Ne5
69
Around now White will start to regret
caving under the pressure to ‘do something’ with
15.g5.
17.Nde2
17.Bf4 Nd6!„ White’s kingside attack is
distant memory, and the fun is just getting started
for Black.;
17.a3 Nd6 18.Nde2 Ndc4 19.Bd4 d5!
20.Bxc4 Nxc4 21.Bxg7 Kxg7 22.exd5
Bxd5 23.Nxd5 Rxd5 24.Rxd5 exd5 25.Qd4+ Qe5
26.Qxe5+ Nxe5 27.Nd4 Rc8 28.c3 Nc6=;
17.Qg3 Nd6„
17...Nc4 18.Bxc4
18.Bd4 d5 19.Bxc4 Bxd4 20.Rxd4 dxc4
21.Rxd8 Rxd8„
18...Qxc4 19.Bd4
19.Rd2 d5 20.Rhd1 a5! 21.exd5 b4 22.b3
Qa6 23.Na4 Rxd5 24.Rxd5 Bxd5
 25.Nb6 Rb8 26.Nxd5 exd5 27.Bd4
(27.Rxd5? Nc7 28.Rd3 Nb5µ)
27...Bxd4 28.Nxd4 Nd6=
19...Bxd4 20.Rxd4

20.Nxd4 Qc5! 21.Rd2 Nd6 22.Nb3 Qxf2

23.Rxf2 Nc4 24.Nc5 Bc6=;
20.Qxd4 Qc7!=
20...Qc5 21.f4 Nd6!„

70
 Appendix (12.0-0 b5 13.Rfe1 d6 14.a3 Rb8 15.Red1
Nc4 16.Bxc4 bxc4 17.Rab1 Bb7 18.Nde2 d5³)
12...b5 13.a3 Bb7 14.Kb1
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6 (14.Ba2 Nc4 15.Bxc4 Qxc4 16.Nb3 a5
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.f3 17.Qd3 Qxd3 18.Rxd3 b4³)
Qc7 10.Qd2 Na5 14...d5!

15.exd5
This may (with further research) turn out to (15.Bxf6 Bxf6 16.exd5 Nxb3 17.Nxb3
be playable (and if my mainline runs into Bxc3 18.bxc3 Bxd5–+)
problems this is a serious fallback to research 15...Nxb3 16.Nxb3 Nxd5µ;
further) but I prefer to continue developing before 11.g4
committing the knight. In the early stages of
developing this system I tried out all kinds of
move orders, and ultimately realized that Black
should keep the knight on c6 so as to deter White’s
Be3-h6 as long as possible (when ...Nc6xd4 would
win a piece).
11.0-0-0

11.Bg5 e6!

11...b5!
(11...e6?! If Black is intent on this 10...Na5
variation then he should prefer 11...b5 instead.
12.h4 (12.0-0-0 b5 13.Bh6 Bxh6 14.Qxh6 b4
15.Nce2 Bb7 16.h4 Nxb3+ 17.axb3 Rfc8∞

The core theme of my system. 12.0-0-0

71
 18.Rd2 (18.c4 bxc3 19.Nxc3 d5³)
18...Bxe4 19.fxe4 Nxe4 20.h5 (20.Rf1 d5µ)
20...g5∞) 12...h5 13.0-0-0 b5 14.Bh6± Again,
Black wishes his knight were on c6 to deter this
move.)
12.g5 Nxb3 13.axb3 b4„
Black is just in time: he will play Nh5 only
after White’s knight moves from c3 (and therefore
cannot reach d5).;
11.0-0 I include this move, because
computers endorse it, but I can’t imagine humans
combining f3 with 0-0. 11...b5 12.a4 b4 13.Nd5
Nxd5 14.Bxd5
72
14...Bb7
(14...Rb8 Black may prefer this. 15.Rac1
e6 16.Ba2 Nc4 17.Bxc4 Qxc4 18.b3 Qc7 19.c3
bxc3 20.Rxc3 Qa5=)
15.Qxb4
(15.Bxb7 Qxb7µ; ¹15.b3 e6 16.Bxb7
Qxb7³)
15...Bxd5 16.exd5 Rab8 17.Qc3 Rfc8µ
Black is calling all the shots. In most openings,
White can get away with a few inaccuracies,
especially with innocent developing moves, but
the price of an inaccuracy is quite high for White
in this system.
11...b5
12.h4!
12.g4 Black generally has three ways of
meeting this move, all intended to prevent White
from playing g4-g5 and Nc3-d5:
1) ...Bc8-b7
 2) ...e7-e6
3) wait for g4-g5 and meet it with ...b5-b4.
Concrete factors determine which of these
is the best in each particular situation. 12...Bb7
13.h4
(13.g5?! Nh5 14.Nd5 Bxd5 15.Bxd5 Rab8³
A sample continuation may be... 16.Kb1
Nc4 17.Bxc4 bxc4 18.c3 Rb7 19.Ka1 Rfb8 20.Rb1
a5 21.Ne2 a4 22.a3 Rb3 23.Ka2 (23.Nc1 Bxc3)
23...e6! 24.Ng3 Nxg3 25.hxg3 R8b5 26.Qh2 Qd6
27.Qxh7+ Kf8–+ White is defenseless against the
combined threats of ...Bg7xc3 and infiltration with
...Qd6-d3 (threatening ...Rxb2+).)
13...Rac8!
An isolated instance where Black can
allow White to play h4-h5.
(13...h5?! Black would ideally like to
prevent White’s h4-h5 but since Black’s knight is
on a5, White now has 14.Bh6!‚ White threatens
Nd4-f5, so the following is forced 14...e6
15.Rhe1! Nxb3+ (15...d6?? 16.Bxe6+–) 16.axb3
73
d6 17.gxh5 Nxh5 18.Bxg7 Kxg7 19.Rg1 Kh7
20.Qg5² Black is on the defensive. Things aren’t
so dire, but we can do better (13...Rac8!).)
14.Kb1
(14.h5?! b4 15.Na4
15...Nxe4!! 16.fxe4 Bxe4 17.hxg6 Nxb3+
18.axb3 fxg6³ 19.Rh2? e5–+)
14...e5! 15.Nde2 d5! 16.g5 Nxb3 17.axb3
d4 18.gxf6 dxe3 19.Qd6 Bh8 20.Qxc7 Rxc7
21.Rd6 Rfc8 22.h5 Rc6 23.Rd7 R6c7 24.Rd6=;
12.Kb1
12...Bb7
(12...Nc4?! This should ‘feel’ wrong;
Black needs to harness more potential energy
before striking. 13.Bxc4 bxc4 14.g4‚)
13.a3
(13.g4 The position is very rich, and Black
has many possibilities. I’ll just show a couple of
them. 13...Rac8 (13...e6 14.h4 (14.Nde2?! d5
15.g5 Nh5 16.exd5 b4 17.Ne4 exd5 18.Bxd5
(18.Nc5? d4! 19.Bxd4 Bxf3 20.Bxg7 Nxg7
21.Nd7 Rad8 22.Rhf1 Nxb3 23.axb3 Rxd7
24.Qxd7 Qxd7 25.Rxd7 Bxe2µ) 18...Rad8 19.Bf4
Qb6 20.Be3 Qb5 21.Bxf7+ Kxf7 22.Nd6+ Rxd6
23.Qxd6 Nc4 24.Qc7+ Kg8 25.Bc5 Na3+ 26.Kc1
Qxe2 27.Bxf8 Qe3+ 28.Rd2 Bxb2+ 29.Kxb2
Qxd2 30.Bh6 Qd4+ 31.Kb3 Qd5+ 32.Kb2 Qd4+
33.Kb3=) 14...d5! 15.e5 (15.g5 Nh5 16.exd5 Nxb3
17.cxb3 b4 18.Ne4 Bxd5„) 15...Nd7 16.f4 Nc4
17.Qe1 (17.Bxc4 dxc4 18.Rhf1 Nc5„) 17...Nc5
18.h5 Nxe3 19.Qxe3 g5!!„

18...e5! This is a thematic push in this line,

made possible by Black holding back the d7 pawn.
19.dxe6 fxe6³;
b) 14.h4 e5! 15.Nde2 Nxb3 16.cxb3
(16.axb3 d5 17.exd5 Nxd5) 16...b4 17.Nd5 Bxd5
18.exd5 e4 19.Rc1 Qb7µ;
13...e6!
(13...Rad8 14.Qf2 Qb8!?„
a) 14...d5 15.Nf5 dxe4 16.Bb6∞;
b) 14...Nc4 15.Bxc4 bxc4 (15...Qxc4
) White has two main options now. 16.Nb3²) 16.Nde2²);
a) 14.g5 Nh5 (14...b4 14.g4 d5!

Things get pretty insane in this line. 15.e5

Fortunately it isn’t necessary for Black to go into (15.g5 Nh5 16.exd5 Nxb3 17.cxb3 Rad8!
this. 15.Nf5! 18.Qc2 (18.dxe6? Bxf3–+) 18...exd5„)
(15.Na4 Nxe4! 16.fxe4 Bxe4 17.Rc1 Nxb3 15...Nd7 16.f4 Nc4 17.Bxc4 dxc4 18.Rhf1
18.axb3 Bxh1 19.Rxh1 f6=) 15...bxc3 16.Nxe7+ Nc5„;
Kh8 17.Qf2 Nh5 18.Bb6 (18.Nxc8 Qxc8) 18...Qe5 12.Bg5 Bb7³;
19.Nxc8 Nxb3 20.axb3 Bxc8=) 15.Nd5 Bxd5 12.Bh6
16.exd5 Nc4 17.Bxc4 bxc4 18.c3

74
 12...Nxb3+ 13.Bh6!
(12...e5?? 13.Nf5+– Beware of this!;
12...Nc4? Too soon, Black needs to harness more This poses Black the most problems when
potential before striking 13.Bxc4 Bxh6 14.Qxh6 the knight is on a5.
bxc4 15.h4+– White is too fast, Black has no 13.g4!? Bb7 14.Bh6‚
counterplay.) (14.g5 Ne8 15.Kb1 Nd6! 16.Nd5 Bxd5
13.Nxb3 17.Bxd5 Rac8³ 18.Qd3 e6 19.Bb3 Ndc4 20.Bc1
(13.axb3 Bxh6 14.Qxh6 b4 15.Nd5 Nxd5 d5! 21.exd5 Rfd8 22.Nxe6 fxe6 23.Qxg6 Na3+
16.exd5 e5! 17.Nf5 gxf5 18.Qg5+ Kh8 19.Qf6+ 24.Ka1 Nxb3+ 25.cxb3 Qc2–+)
Kg8=) 13...Nxb3+ 14.axb3
13...Bxh6 14.Qxh6 Bb7„
14.Nxb3 Bxh6 15.Qxh6 Bb7 16.a3 Rac8
17.Kb1 a5„
14...Bb7 15.g4 e6 16.b4!±

Black will continue ...Rfc8, ...a5, ...b4, etc.

12...h5

Such positions are to be avoided: White

dominates on all sides of the board, which will be
evident after Nd4-b3-a5(c5).

75
 CHAPTER 4

MAROCZY BIND: BREYER

VARIATION

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5
We have here the starting position of the
Maroczy Bind, named after Hungarian master
Geza Maroczy (1870–1951). Maroczy was one of
the best players in the world in the early 1900s,
and was even set to play a World Championship
match with Emanuel Lasker in 1906, but for
various political reasons the match did not end up
taking place. He was an engineer by training, and a
financial auditor by profession. Capablanca, near
the end of his life in the early 1940s, said of
Maroczy that “with the exception of Botvinnik and
Keres, Maróczy in his time was superior to all the
other young masters of today.” Maroczy
popularized White’s setup in his capacity as a
chess journalist, but there are no published games
Contents of his where he played the system with White. In
fact, the first recorded game in the Maroczy Bind
is Swiderski-Maroczy, 1904, where Maroczy
1. 6.Nc2 Nf6 defended the Black side of the Accelerated Dragon
2. 6.Nc2 e6 against the (to be named) Maroczy Bind. In order
3. 10.Be2 to combat the Maroczy Bind, Black must follow a
4. 10.Nb5 0-0?! dark-squares strategy. Notice that in asserting the
5. 10.Nb5 d6!N 11.sidelines central c4/e4 ‘clamp’, White has forever denied
6. 10.Nb5 d6!N 11.Qd2 the (dark) d4-square protection from a pawn. We
7. 10.Bd3 d6 will see in the subchapter 3 that the Breyer
Appendix. 10.Bd3 0-0 Variation is based on seizing this square from
White’s grip. In keeping with the dark-squares
strategy, exchanging dark-squared bishops is in
1 general congenial to Black. In fact, GM Tiviakov
once highlighted the importance of this exchange
by playing the highly creative, though slightly
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7 dubious, 1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6
5.c4 5.c4 Bh6!?.
5...Nc6 6.Nc2

76
 This is not the most challenging line for
Black, but since it has been played by Keres, Tal,
Korchnoi, and Carlsen, it has earned our serious
attention. White usually waits for Black to play
...Bc8-d7 before retreating this Knight, because
Black’s two most common ways of countering
Nc2 are either a) ...Nf6-d7-c5, often capturing on
c3 with the g7 bishop (an option made available by
White’s departure from d4), or b) ...Bc8-e6 (which
is not recommended when White’s knight is on
d4) and quickly targeting the c4 pawn with
...Ra8-c8, ...Nc6-e5, etc. The latter of these plans
will be covered when we discuss the ‘Main Line’
Maroczy. Here I will go over two additional
systems for the reader who wishes to take
advantage of White’s premature retreat.
6...Nf6 7.Nc3 0-0 8.Be2 b6
The double-fianchetto is particularly adept
at meeting White’s Nd4-c2. With the knight on d4,
a reliable way to meet Black’s double fianchetto is
with Nd4xc6, which is not possible here.
9.0-0 Bb7
77
Black intends to continue with the at first
glance unbelievable, but in this book thematic,
...Nf6-e8-d6, followed by ...f7-f5.
10.f3
10.Bf4 This deters Black’s ...Ne8-d6 idea
but after ...Nf6-e8 Black may instead gain a tempo
on the f4-bishop after ...Ne8-c7-e6. Also, White’s
f4-bishop is under X-ray attack from Black’s f8
rook after an eventual ...f7-f5. 10...Rc8 11.Ne3
Ne8 12.f3
(12.Ng4?? h5 13.Ne3? e5–+
Herman-Panjwani, New York 2014)
12...Nc7 13.Ned5 Ne6 14.Be3 f5 15.exf5
Rxf5„
Black will continue with ...Ne6-d4,
...e7-e6, and eventually ...d7-d5.;
10.Bg5 Not the most popular, but quite
possibly White’s best. 10...Ne8!
(10...Rc8?! Black should not delay
...Nf6-e8, because that allows White to coordinate
in the center and render the plan too slow. 11.f3 h6
12.Be3 Kh7 13.Qe1 Ne8?! 14.Rd1 Nd6?! 15.Qd2!
(15.Kh1 f5 16.c5 bxc5 17.Bxc5 fxe4 18.Bxd6 exf3
19.Bxf3 exd6„ 0–1 (35) Sadorra,J
(2431)-Jobava,B (2664) Al Ain UAE 2008) 15...f5
16.exf5 Nxf5 17.Qxd7 Nxe3 18.Nxe3±)
11.Qd2
(11.f3 Bxc3 (11...Nd6 12.Qxd6! exd6
13.Bxd8 Raxd8 14.Rad1² Black certainly has
compensation for the weakened structure, because
White is without his prized dark-squared bishop,
but I think Black’s compensation will be at best
enough for a draw, and as a general rule it is worth
avoiding positions where your opponent is playing
for two results.) 12.bxc3 Ng7„
Black will play in accordance with the
system of Nimzowitsch: first blockade White’s
doubled c-pawns with ...d7-d6, ...Ng7-e6-c5, then
pile up on the weak c4-pawn with ...Ra8-c8,
...Nc6-a5, ...Bb7-a6.)
11...Nd6 12.f3
78
12...f5!
(12...f6 It is instructive to see how Black
ends up worse after this move. 13.Bf4 (13.Bh6
Bxh6 14.Qxh6 Ne5 15.b3 g5„) 13...Ne5 14.Ne3 f5
15.exf5 Nxf5 16.Nxf5 Rxf5 17.Be3 Rc8 18.b3
Black cannot allow White to coordinate like this,
because Black has made anti-positional moves and
he can only get away with them if he plays
energetically enough to prevent White from
‘achieving harmony’. Therefore, Black lives and
dies by the power of his counterplay: here,
unfortunately, Black has capitulated control over
the center, and is consequently unable to achieve
the desired ...d7-d5. 18...Nf7 19.Rad1 Nd6
20.Bd4²)
13.exf5
Both recaptures on f5 are interesting, but
my analysis indicates that 13...Nxf5 is the better of
the two. 13...Nxf5
(13...Rxf5?! 14.Bh6! This move, which is
all too easy for White to find, makes life difficult
for Black. From a broader perspective this should
not surprise us: by exchanging pieces White dulls
Black’s counterplay, which is Black’s only
compensation for his hanging central pawns and
multiple weaknesses. (14.Be3 Qf8! 15.Rad1 Rf7
16.c5 Nf5 17.cxb6 e6 18.bxa7 Nxe3 19.Qxe3
Nxa7©) 14...Qf8 15.Bxg7 Qxg7 16.b3 Raf8
17.Rad1 g5 In a practical game Black can
probably get away with playing like this, but I
cannot recommend this as something to aim for
from the outset. 18.Bd3 Rf4 19.Ne4 Ne5 20.Nxd6
exd6 21.Be2! R8f6 22.Ne3²)
14.Rad1 Rc8! 15.b3
(15.Qxd7? Bd4+ 16.Rxd4 Nfxd4 17.Qxb7
Nxc2³)
15...h6! Black guarantees himself the two
bishops, if he desires. 16.Bf4 g5
Irrespective of where White moves his
bishop, Black will continue with ...e7-e6 and
...d7-d5. It is amazing that Black can get away
with this central break seeing as White currently
has four pieces directly targeting at the d5 square
and Black has none! 17.Bg3
(17.Be3 e6 (17...d5!? 18.Nxd5 (18.cxd5??
Bxc3 19.Qxc3 Ncd4–+) 18...e6 19.Ndb4 (19.Nc3
Qf6 20.Nb5 Rcd8 21.Qc1 a6 22.Nba3 Nxe3
23.Qxe3 Nd4 24.Nxd4 Qxd4!=) 19...Nxb4
20.Qxb4 Qc7 21.Bf2 Be5 22.h3 Ng3 23.Bxg3
Bxg3∞) 18.Bd3 (18.Nb5 d5! 19.cxd5 exd5
20.Qxd5+? Qxd5 21.Rxd5 Na5 22.Rd2 Rxc2
23.Rxc2 Nxe3–+) 18...d5! 19.Bxf5 Rxf5 20.cxd5
exd5 White’s position is slightly more pleasant but
Black shouldn’t have much difficulty holding.
21.Nd4 (21.Nxd5 Qd6©) 21...Nxd4 22.Bxd4
Bxd4+ 23.Qxd4 Qf6
79
Black will have enough activity to
compensate for his pawn deficiency. It is not
worth memorizing an exact path to equality in the
ensuing variations; there are multiple paths.
24.Qxf6 (24.Nb5 Ba6 25.Qa4 Bxb5 26.Qxb5 d4
27.Qd3 Rd5=) 24...Rxf6 25.Nxd5 Re6 26.Nb4 a5
27.Nd3 Ba6 28.Rfe1 Rxe1+ 29.Nxe1 Re8 30.Nc2
Re2 31.Rd8+ Kf7 32.Nd4 Rxa2 33.Rd6 Ra1+
34.Kf2 Ra2+ 35.Ke3 h5 36.Rxb6 Bf1=)
17...e6
18.Nb5
(18.Bf2 d5!! 19.cxd5? Bxc3 20.Qxc3
Ncd4µ)
18...d5! Black will be fine after a long
forcing series of exchanges. 19.cxd5 exd5
20.Qxd5+ Qxd5 21.Rxd5 Ncd4 22.Rxd4 Rxc2
23.Bc4+ Kh8 24.Rd3 g4! 25.Nd6 Nxd6 26.Bxd6
Re8
27.a4
(27.Bb4 Preventing ...Re8-e2 on account of
Rd3-d8+. 27...gxf3 28.gxf3 Rxa2=)
 27...Ree2 The following variation is
remarkable: 28.Kh1! Rxg2 29.Bd5 Ba6 30.Re3!
Bxf1 31.Re8+
31...Bf8
(31...Kh7 32.Be4#)
32.Rxf8+ Kg7 33.Rg8+ Kf6 34.Rf8+ Kg7
35.Rg8+ Kh7 36.Be5!! gxf3 37.Rh8+ Kg6
38.Be4+ Kh5 39.Bxf3+ Kg5 40.Bxg2 Bxg2+
41.Kg1=;
10.Rb1 Ne8 11.Qd2 Nd6 12.f3 f5 13.Nb5
fxe4 14.Nxd6 exd6 15.Qxd6 Ne5 16.f4 Nd3
17.Bxd3 exd3 18.Qxd3 Re8© 0–1 (38)
Tseshkovsky,V (2555) -Gufeld,E (2535) Vilnius
1975.
10...Ne8
11.Be3
11.Bd2 Played in order that Black will not
capture on c3, but to seek an advantage White
should allow that. 11...Nd6 (…...f5) 12.Rc1 f5
80
13.exf5 Nxf5 14.Kh1 e6 15.Qe1
15...Kh8
(15...d5!? 16.cxd5 exd5³)
16.Ne3 Nfd4
(16...Ncd4 17.Bd3 Nxe3 18.Bxe3 d5?!
19.cxd5 exd5 20.Rd1 Nf5 21.Bxf5 Rxf5 22.Bd4²
1–0 (46) Hebert,J-Panjwani,R Montreal 2013)
17.b3 d5!³
11...Nd6
A significant branching point for this
variation. White’s attempts in the database are not
convincing, but I have found a new move
(12.Nd4) which does present Black with some
problems. Fortunately, they are not
insurmountable.
11...Bxc3!? 12.bxc3 d6 13.Bh6 Ng7 14.f4
Re8 15.Bg4∞
12.Nd4N
As I mentioned before, White’s knight
belongs on d4 in these double-fianchetto systems,
especially in the cases where Black’s play revolves
around ...f7-f5. White intends on eliminating
Black’s pesky d6 knight with Nd4-b5.
12.Qd2 f5
 16.f4 e6 17.Be5 f6 18.Bd6 Rf7= Black can play
around the d6 bishop, and always get rid of it with
...Bb7-a8 and ...Na5-b7 if necessary.
13...f5! 14.exf5 Nxd4

14...gxf5 15.f4 e6 16.Nxc6 Bxc6 17.Qxd6

Bxc3 18.Rad1²
15.Bxd4 Bxd4+ 16.Qxd4 Nxf5 17.Qd2
e5!

13.c5
(13.exf5 Nxf5 14.Bf2 Bh6 15.Qe1 (15.Qd1
e6 16.g4 Nfe7 17.Bg3 d5 18.cxd5 Nxd5 19.Nxd5
exd5„) 15...e6 16.Rd1 Qe7 17.g4 Ng7 18.Bg3
Bf4=)
13...bxc5 14.Bxc5 fxe4 15.fxe4 Rxf1+
16.Rxf1 Ne5 17.Bxd6 Qb6+ 18.Kh1 Qxd6=;
12.c5 bxc5 13.Bxc5 f5 14.Bxd6 exd6
15.Kh1 Qh4 16.g3 Qh3 17.exf5 Be5 18.Rg1
Ne7!„;
12.Nb5 Bxb2 13.Rb1 Be5 14.f4 Bg7 18.Rad1 Nd4
15.Nxd6 exd6 16.Qxd6 Re8 17.Qd3 Na5 18.Bf3
Rc8³ 0–1 (37) Minasian,A (2478)-Aronian,L The position is equal, even if White has the
(2528) Ohrid 2001. ‘better half’ of it. Things will eventually fizzle out;
12...Rc8! the following is a sample.
19.Nb5 Nxb5 20.cxb5 d5 21.f4 Qe7
22.fxe5 Rxf1+ 23.Rxf1 Qxe5 24.Bf3 Re8 25.Rd1
Kg7=

12...f5?! 13.Ndb5 Nxb5 14.cxb5 Na5

15.Bd4²
13.b3
Black should stick to waiting tactics — let
13.Ndb5 Nxb5 14.cxb5 Na5 15.Bd4 Bh6 White try to make progress. At the right moment
81
Black may ditch the d5-pawn to arrive at a drawn 5.c4 Nc6 6.Nc2 e6! 7.Be2
rook endgame.
2 7.Nc3 Nge7 8.Bf4 This is the most obvious
way to counter Black’s setup — planting a bishop
on d6. It turns out, however, that Black need not
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7 be bothered by such an intruder.
5.c4 Nc6 6.Nc2 e6! a) 8.h4

This move is almost a novelty, but (to my

dismay) not quite: it has been played a mere four
times (and with different intentions to ours) in the
recorded history of our game.

8...f5 (8...0-0 9.h5 d5!? Further research is

warranted here.) 9.exf5 (9.h5 fxe4 10.Nxe4 d5
11.Nc3 0-0„) 9...Nxf5 10.h5 0-0 11.hxg6 hxg6
12.Bf4 (12.Qg4 Ne5 13.Qh3? Kf7! 14.Ne4 Rh8
15.Ng5+ Kg8–+; 12.Ne3 Qb6„) 12...Qb6 13.Rb1
The idea of this move occurred to me while Nce7!
I was analyzing the ...Ne8-d6 maneuver in the
6...Nf6 line, where Black often plays ...f7-f5 and
recaptures on f5 with the knight. I wondered
whether Black could reach similar positions and
save some time by taking the route Ng8-e7-f5
rather than Ng8-f6-e8-d6-f5. I was also aware of
similar ideas with colors reversed in the English
opening, such as after 1.c4 e5 2.g3 Nf6 3.Bg2 d5
4.cxd5 Nxd5 5.Nc3 Nb6 6.e3!?
or 1.c4 c5 2.Nc3 Nf6 3.g3 d5 4.cxd5 Nxd5
5.Bg2 Nc7 6.e3!?. Black’s basic idea is to go for a
quick ...f7-f5, and then recapture on f5 with the
knight, from where it controls the important d4
square and also often pressurizes White’s e3
bishop. White can take one of (at least) two Black can ‘premove’ ...d7-d5 here; with
strategies: either try to develop normally and hope White’s king in the center Black shouldn’t hesitate
that Black’s offbeat play will lead to static to sacrifice a pawn to open files. 14.Ne3
positional weaknesses in the middlegame, or a1) 14.Be2 d5! 15.cxd5 exd5 16.Nxd5?
White can aim for a refutation of the system with Nxd5 17.Qxd5+ Be6µ;
an early h2-h4 or Bc1–f4-d6. According to my a2) 14.Qd2 d5! 15.g4 Nd6 16.Be3 Qc7
analysis, Black equalizes in all variations. 17.Bh6 (17.cxd5 Nc4 18.Bxc4 Qxc4„) 17...Nf7
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
82
18.Bxg7 Kxg7=;
14...d5! 15.cxd5 exd5 16.Ncxd5 Nxd5
17.Qxd5+ Be6
18.Qxe6+ Qxe6 19.Bc4 Qxc4 20.Nxc4
Rae8+ 21.Ne3 Re4 22.Bg5 Bf6 23.Bxf6 Rxf6
24.Rh3 g5=;
b) 8.Bg5 0-0 9.Qd2 Qa5 10.Bh6 (10.Be2
d5!„) 10...Bxh6 11.Qxh6 f5!„;
8...0-0 9.Bd6
9...f5 White has taken time to occupy d6;
as such, he is behind in development so Black
should not delay this counter-punch.
(9...b6?! 10.h4 Bb7 11.h5 f5?! 12.hxg6
hxg6 13.Qf3 fxe4 14.Qh3 Kf7 15.Nxe4 Rh8
16.Ng5+ Kg8 17.Qxh8+ Bxh8 18.Rxh8+ Kxh8
19.Nf7+ Kg7 20.Nxd8 Rxd8 21.0-0-0±) 10.exf5
(10.Bd3 Rf7 11.0-0 b6 12.Qd2 Bb7 13.Rae1 Rc8
14.b3 Ne5 15.Bxe5 Bxe5 16.exf5 Bxc3 17.Qxc3
gxf5=)
10...Rxf5 Although Black would normally
83
prefer to take on f5 with the knight, the knight will
find its way to f5 anyway after ...Rf5-f7. 11.Bd3
(11.Ne3 Rf7 12.h4 Nd4 13.h5 Nef5
14.hxg6 hxg6 15.Nxf5 Nxf5 16.Bf4
16...b5!! A shot out of the blue. Black can
afford to sacrifice a pawn in order to accelerate his
development because White’s king is still
awkwardly uncastled. 17.Nxb5 (17.cxb5 d5©)
17...Bb7 18.Qd2 (18.Qg4 Bxb2 19.Rd1 Qf6„)
18...Qb6 19.Rh3 Raf8 20.0-0-0 d5!„)
11...Rf7 12.0-0 b6 13.Ne3 Bb7 14.Re1 Rc8
15.Rc1 Nf5 16.Nxf5 gxf5 Routine development
for both sides has led to a dynamically equal
position.
Black looks forward to occupying the
g-file with his rook, which combines nicely with
pressure from the b7-bishop. 17.Be2
(17.Re3 Qf6 18.Rg3 A little knight sortie
solidifies Black’s kingside and calls into question
the placement of White’s g3-rook. 18...Ne5! 19.b3
Ng6!=)
17...Nd4 18.Bh5 Qg5!
19.g3
(19.Bxf7+ Kxf7 20.Bg3 f4 21.Qd2 Bxg2!
22.Qxf4+ Qxf4 23.Bxf4 Bb7 24.Kf1 Rxc4= Two
bishops and two central pawns are more than
enough compensation for Black.)
19...Rxc4 20.b3?
(20.h4 Qd8 21.b3 Rc6 22.Bxf7+ Kxf7
23.Qh5+ Kg8 24.Be5 Bxe5 25.Rxe5 b5!©)
20...Bf3!! 21.Bxf3 Qxc1!! 22.Qxc1 Nxf3+
23.Kg2 Rxc3 24.Qd1 Nxe1+ 25.Qxe1µ;
7.h4 Nge7 8.h5 d5! 9.h6 Bf6³
7...Nge7 8.Nc3
84
8...f5!
As is so often the case in this opening,
Black’s play is contrary to basic chess principles;
usually one is supposed to castle before such
breaks. Then again, White’s king isn’t castled
either.
The only top-level game continued as
follows: 8...0-0 9.0-0 f5 10.exf5 gxf5?! This
misses the point; Black should recapture with the
knight! (10...Nxf5=) 11.f4 d5! 12.cxd5 exd5
13.Rf3 Be6 14.Bd3 Rc8 15.Rg3 Rf6! 16.Bd2 Rg6
17.Qe1 Bf7 18.Kh1 Qd7 19.Qf2 Re8 20.a3 Nc8!
21.Re1 Rxe1+ 22.Bxe1 Nd6µ 1/2–1/2 (56)
Yanofsky,D-Stoltz,G Karlovy Vary 1948.
9.h4
When I thought of this system, I showed it
to a 2600+ rated friend, and he claimed he was
confident he could refute this line over the board.
This was his first attempt (of many unsuccessful
attempts!).
9.exf5 Nxf5 10.0-0 0-0 11.Bf4 b6 Both of
Black’s bishops occupy the longest diagonals on
the board. 12.Qd2 Bb7 13.Rad1 Ncd4 14.Nxd4
Nxd4=;
9.0-0 fxe4 10.Nxe4 d5 11.Nc3 0-0=
9...fxe4 10.h5
10.Nxe4 d5„;
10.c5 b6 11.Nxe4 0-0 12.h5 bxc5 13.hxg6
Nxg6µ
10...d5 11.hxg6 hxg6 12.Rxh8+ Bxh8
White has compensation for the sacrificed
pawn but only enough for equality, and that is with
perfect play from here on out.
13.Bg5! Qa5 14.Qd2 Bd7 15.0-0-0 Bxc3
16.Qxc3 Qxc3 17.bxc3 0-0-0 18.Ne3 Rf8=

85
 3
The Breyer variation of the Maroczy is
characterized by Black’s 7...Ng4. Unlike the
Maroczy Bind (5.c4), which was never played by
its namesake Geza Maroczy, the Breyer variation
was indeed first played by Gyula Breyer in
Kostic-Breyer, Gothenburg 1920. It has since been
entrusted for important games by World
Champions Botvinnik (Smyslov-Botvinnik,
Alekhine Memorial 1956) and Kasparov
(Ribli-Kasparov, World Cup 1988).
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3
7.f3 Qb6!
8.Nf5 Qxb2 9.Nxg7+ Kf8 10.Nd2
(10.Bh6?? Qxa1–+)
10...Kxg7 11.c5 Rd8 12.Rb1 Qe5 13.Nc4
Qe6³ Black will free his position with either
...d7-d5 or ...b7-b5.
7...Ng4
86
8.Qxg4
8.Nxc6?! This is what they played in
Breyer’s time. 8...Nxe3 9.Nxd8 Nxd1 10.Kxd1
Kxd8³
8...Nxd4
8...Bxd4? It is reasonable to ask why this
move is not played more often, given that Black
usually seeks to exchange dark-squared bishops in
the Marozy. True, in the middlegame — after both
sides have developed — the trade of dark squared
bishops is generally desirable for Black, but here
Black purchases this exchange at the cost of
precious time. 9.Bxd4 Nxd4 10.0-0-0! Nc6
11.Qg3! 0-0 12.h4± White’s attack is
overwhelming.
9.Qd1
9.0-0-0 Nobody plays this. 9...Nc6 10.Nd5
e6!
(10...0-0 11.Qh4‚)
11.Nc3 Qa5„
9...e5
 Bent Larsen used to prefer 9...Ne6, but I
think 9...e5 offers more chances for counterplay
based on Black’s occupation of the center.
Computers generally tend to prefer White in the
Breyer, but from my experience even 2600+ GMs
find it very uncomfortable to ‘play around’
Black’s d4 knight, which (invariably) leads them
to mishandle White’s position. This line has a very
Sveshnikov Sicilian feel to it. Fischer taught us
that “you have to give squares to get squares”;
Black has opted to relinquish control over the light
squares (especially d5) in order to stake a claim
over the all-important d4 square. This is the first
significant branching point for White: he will
either develop routinely with Bd3, 0-0, Qd2, etc.,
or he will attempt to immediately take advantage
of Black’s weaknesses with 10.Nb5. We consider
these options in the coming chapters, but we begin
with an underrated sideline.
10.Be2
If this move is followed by routine
development, then it lacks independent value,
since White’s bishop is better placed on d3.
However, there are some idiosyncrasies to this line
if White combines Bf1–e2 with Nc3-b5.
10...0-0
10...d6 The reader can make life easier for
themselves by playing 10...d6 in response to any
of White’s major options, but since he can get
away with it, it is at least symbolically more
flexible for Black to castle first.
11.Nb5!
87
A tricky move, not least of all because the
response recommended by the computer at even
high depths (11...Qh4) is a blunder due to a
brilliant double-rook sacrifice demonstrated by
former Women’s World Champion Nona
Gaprindashvili more than 40 years ago!
11.0-0 d6 12.Qd2
(12.Nb5 Nxb5 13.cxb5 Be6 14.Qd2
(14.Bf3?! Bc4=) 14...f5 15.Bg5 Bf6 16.Bxf6 Qxf6
17.Qxd6 Rfd8 18.Qc7 Rd7 19.Qc5 fxe4 20.Bc4
Rd4 21.Bxe6+ Qxe6=)
12...Be6= White can obtain this position
with his bishop more appropriately placed on d3.
In fact, my computer recommends that White play
13.Bd3 here.
(12...f5?!
Black shouldn’t rush ...f7-f5; the following
is just a blitz game I played over the internet
which gives an illutration of the sort of
counterplay Black can generate. 13.f3 (13.exf5
Nxe2+ 14.Qxe2 gxf5 15.Rfd1 Qe7 16.c5!±) 13...f4
14.Bf2 g5 15.Nb5?! Nxe2+ 16.Qxe2 a6 17.Nc3 g4
18.fxg4 Qg5 19.h3 h5„)
11...Nxb5!

Black is well advised to transpose into the

variations after 11.0-0 d6.
11...Qh4? It is a pity this doesn’t work for
Black (on account of 12.Nxd4!), because many of
the ensuing variations illustrate the elastic
potential for Blacks position to suddenly ‘come to
life’. 12.Nxd4!
a) 12.Qd3 d5!! 13.cxd5 (13.exd5? Bf5µ)
13...Nxe2 14.Kxe2 (14.Qxe2 Qxe4=) 14...f5 15.f3
fxe4 16.Qxe4 Qf6!„;
b) 12.Nd6 Qe7 13.Nxc8 Rfxc8 14.0-0 Qb4
15.b3 a5 16.Rb1 a4 17.Bxd4 (17.Bd3 axb3
18.Bxd4 exd4 19.Rxb3 Qd6= See 17.Bxd4 for
transposition.) 17...exd4 18.Bd3 axb3 19.Rxb3
Qd6 20.Qb1 Rcb8 21.f4 Qc7 22.e5 Ra5=;
c) 12.Bf3? d5!!
12...Qxe4!! (12...Rb8!? 13.Bd3 (13.Qd3?
f5 14.exf5 d5!!µ) 13...b6 (13...d6 14.0-0 f5 15.f3²)
14.0-0 Bb7 15.Nd5 (15.f3 f5 16.exf5?! Nxf5
17.Bxf5 gxf5 18.Qxd7 Qxc4 19.Ne6 Rf7 20.Qd6
Re8 21.Nd8 Bf8 22.Qd2 Bb4 23.Qd1 Rg7
24.Nxb7 Rxb7=) 15...f5 16.Bxd4 Bxd5 17.cxd5
fxe4 18.g3 Qf6 19.Bxe4 exd4 20.Qd3 White’s
position is more pleasant, but Black can shift
around and ask White to prove an advantage —
most likely things will fizzle out into a draw.)
13.Bxd4 (13.Nxa8 Qxg2 14.Rf1 d5! 15.cxd5 Bf5
16.Bxd4 exd4 17.Nc7 Qh3© Black is a full rook
down but his attack is overwhelming.; 13.Bf3
Nc2+ 14.Kd2 Nxe3 15.Bxe4 Nxd1 16.Raxd1
Rb8=) 13...exd4 14.f3 (14.Nxa8 Qxg2 15.Rf1
Qxh2µ After Black eats up the a8 knight with
...Qh2-b8, he will be materially and positionally
winning.) 14...Qe5 15.Nxa8

13.exd5 Nxf3+ 14.Qxf3 Qxc4³;

d) 12.Nc7

15...Bh6!! (15...d5 16.0-0 (16.cxd5 Bf5

17.0-0 d3 18.Bxd3 Qd4+ 19.Rf2 Bxd3©) 16...Bf5

88
17.Kh1! Rxa8 18.f4²) 16.0-0 (16.Kf1 Be3©)
16...Bf4 17.g3 Bxg3 18.hxg3 Qxg3+ 19.Kh1 Qh3+
20.Kg1=;
12...exd4 13.Bxd4 Qxe4 14.Bxg7 Qxg2

(14...Kxg7 15.0-0±)
15.Qd4!! Engines have difficulty seeing
this brilliant motive several moves in advance, but
humans, especially after Anderssen-Kieseritzky
1851 (better known as “The Immortal Game”),
know to look for such ideas. 15...Qxh1+ 16.Kd2
Qxa1 17.Qf6! 1–0 (17)
Gaprindashvili,N-Servaty,R Dortmund 1974.
12.cxb5 d6 13.0-0 Be6=

We have transposed to 11.0-0 d6 12.Nb5,

which we have already seen is fine for Black.

89
 4

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Nb5

In the 1960s and 1970s, thanks to some

masterpieces by World Champions Smyslov and
Tal, this move was thought to be a near refutation
of Black’s setup. Over the years, preference of top
players has vacillated between 10.Nb5 and
10.Bd3, their popularities remaining about equal to
this day.
11...Re8 12.Nd6 Re6 13.c5²;
11...d6 12.Nxd4 exd4 13.Bxd4 Qh4
14.Bxg7 Qxe4+ 15.Be2 Kxg7 16.0-0±;
11...d5 12.cxd5 Qh4 13.Nc3 Bg4 14.Bg5±;
11...Nxb5 12.cxb5 d6
(12...Qh4 13.Bd3 Qg4 14.0-0 b6 15.Bc4
Bb7 16.f3+–)
13.Bc4²
12.Nc3!N

It may appear odd to retreat this knight to

10...0-0?!
This is what everyone plays, but the
positions Black tends to get are not inspiring, and
White is scoring quite well (over 60%) after this
move in the database. This is why I recommend
instead the powerful novelty 10...d6! in the next
subchapter.
11.Qd2 Qh4!
where it was just a couple of moves ago. White’s
idea is to castle queenside and pawn storm Black’s
king — punishment for castling too early, at least
compared to the line I recommend in the next
chapter. Black’s queen is misplaced on h4, and if it
retreats to d8 then White and Black’s lost tempi
‘cancel out’.
12.Bd3 d5!!

Definitely Black’s best try, but due to the

novelty I recommend for White next move, Black
is left struggling to equalize.

13.exd5

90
 (13.cxd5 Nxb5 14.Bxb5 Qxe4 15.0-0 Rd8
16.Rfd1 Bd7=)
13...Bh3!! 14.Bxd4 exd4 15.gxh3
(15.0-0 Bxg2 16.Kxg2 Qg4+ 17.Kh1 Qf3+
18.Kg1=)
15...a6 16.Nd6
(16.Na3 Bh6 17.Qc2 Qxh3©)
16...Qe7+ 17.Ne4 f5 18.0-0-0 fxe4
19.Rhe1 b5 20.Rxe4 Qc7©
12...d6

12...Qd8 13.h4 h5 14.Ne2 Nc6 15.g4 hxg4

16.h5‚
13.0-0-0 Be6

13...Qd8 14.h4 h5 15.Ne2 Nc6 16.Qxd6

Qa5 17.Nc3 Nd4 18.Qa3 Qxa3 19.bxa3²
14.Bg5 Qh5 15.Kb1 f5 16.f3 Rfe8 17.Be3²

Black’s queen is miserable on h5.

91
 5 presumably 10...d6 was never played because it
was thought to lose a pawn in this way.
11.Be2!?
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Nb5

It is remarkable that in the sixty year

history of this position, which has been defended
by such champions as Kasparov and Korchnoi, not
a single game is in the database after 10...d6!. The
following is arguably the most powerful novelty in
this book; I believe that the ensuing variations
demonstrate the superiority of 10.Bd3 over
10.Nb5.

11...Qa5+!
(11...Nxb5?! This transposes to an inferior
version of our main line. 12.cxb5 Be6 13.0-0 0-0
(13...d5 14.exd5 Qxd5 15.Qxd5 Bxd5 16.Rfd1
Be6 17.Bf3+–) 14.Qd2 d5 15.exd5 Qxd5 16.Qxd5
Bxd5 This is a position from our main line except
there Black plays ...e5-e4 instead of ...0-0. 17.Rfd1
Be6 18.Bf3²)
12.Bd2
(12.Qd2

10...d6!N

I refer to this as a novelty, even though (as

we will see) I have played this move twice (and
won both times!) in tournament play — in
unpublished games against WGM Nemcova
(2350) and IM Gorovets (2550). Black saves a
crucial tempo by delaying ...0-0 which allows him
to generate central counterplay one move sooner.
This does not violate the laws of opening play;
after all, White is not castled either. It is actually
quite in line with the advice given by American
legend GM William Lombardy in his recent book: 12...Qxd2+ (12...Nc2+ 13.Kd1 Nxe3+
“Not only should one not rush to castle, but should 14.fxe3 Qxd2+ 15.Kxd2 0-0 16.Rhd1 Bh6!
delay that passive maneuver for as long as good 17.Nxd6 Rd8 18.c5 b6 19.b4 Be6 20.a3 a5„)
judgment relates that there are more urgent, if only 13.Kxd2 (13.Bxd2 Nxb5 14.cxb5 Be6 15.0-0 f5
slightly better, tasks to accomplish”. 16.Bb4 0-0-0 17.Rfd1 d5=) 13...Nxb5 14.cxb5
11.Nxd4 Be6 15.Bf3 (15.Rhc1 d5 16.exd5 Bxd5 17.b6 axb6
18.Bb5+ Ke7 19.Rc7+ Ke6 20.Ke2 e4µ This was a
This is of course the first line to check: game I played online.) 15...0-0 16.Rhc1 a6
92
17.bxa6 (17.b6? f5! 18.exf5 gxf5 19.Bxb7 Rab8
20.Bxa6 f4µ) 17...Rxa6 18.Rc7 Rb8= White’s
occupation of the 7th rank is short-lived; Black
will continue ...Bg7-f6-d8 with equality.)
12...Qb6

14.g3
(14.h3 0-0 15.Qc1 f6 (…...g5) 16.0-0
(16.g3 Bxg3 17.fxg3 Qxg3+ 18.Ke2 Qg2+–+;
16.c5 Be6„) 16...Bxh3 17.f4 (17.gxh3 Qxh3–+)
17...Qg3 18.Rf2 Bxg2! 19.Rxg2 Qxd3 20.fxe5
13.Nxd4 dxe5 21.Bh6 Rf7 22.Qc2 Qxc2 23.Rxc2 f5„
(13.0-0 0-0 14.Nxd4 exd4 15.Qb3 Be6
16.Bd3 Qc7 transposes to 13.Nxd4.; 13.Be3 a6
14.Nxd4 exd4 15.Bc1 f5„)
13...exd4 14.Qb3 0-0 15.Bd3 Be6 16.0-0
Qc7 17.Qa3

This will most likely be drawn but I’d

rather be Black than White.)
14...Qh3 15.Bf1 Qe6 16.Qb3
(16.f4 Bg7 17.Bd3 Qh3=)
16...0-0 17.Bg2 Bd7 18.0-0
17...f5 18.exf5 Bxf5 19.Bxf5 Rxf5= Black
will eventually undouble his pawns with ...d6-d5.
11...exd4 12.Bxd4

12.Bd2 White cannot expect an advantage

playing like this; Black has many ways to
equalize. 12...Qh4
(12...0-0!? 13.Bd3 f5 14.0-0 Be6 15.Qe2
Qc7 16.b3 Rae8 17.Rae1 Bd7=)
13.Bd3 Be5
93
 14.g3 Be6 15.Kf3
(15.b3 0-0 16.Bg2 b5!„)
15...f5 16.exf5 Bxf5 17.Kg2 Be4+ 18.f3
Bc6 19.Be2 0-0©
14...f5!

18...b5!„;
12.Bf4?! f5! 13.exf5 0-0³
12...Qa5+ 13.Ke2 Be5!

14...Be6 15.Kf2 0-0 16.Be2 f5 17.b4 Qc7

18.Bxe5 dxe5 19.exf5 Rxf5 20.c5 Rd8 21.Qc2 Rf4
22.Qc3 e4 23.Qe3²;
14...0-0 15.Kf2 Be6 16.Be2 f5 17.exf5
Rxf5 18.Bxe5 Qxe5 19.Qd2 Qc5+ 20.Qe3 Bxc4
21.Bxc4+ Qxc4=
15.Qd3!

15.exf5 Bxf5 16.Kf2 0-0-0

(16...0-0!?)
17.Rc1
(17.Be2? Bc2!³)
Our compensation becomes of a long-term 17...Rhe8 18.Be2 Kb8 19.Rc3 d5„
nature with this move. White cannot exchange 15...fxe4 16.Qxe4
bishops without returning his extra pawn, so he
has to look for ways to disentangle his pieces —
starting with his king on e2. I suspect most players
will not play the most critical lines for White here
(it can’t be easy to if you have never seen this
position before), so Black may pragmatically opt
to just remember 13...Be5! and figure things out
over the board from here.
14.f3

According to my research this is clearly

White’s best try for an advantage.
14.Bxe5 Qxe5
(14...dxe5!?©)
15.Qd2 Qxe4+ 16.Qe3 Qe7=;
94
 16...Be6!

16...Bf5?! It’s too bad this doesn’t quite

work because of 17.Qe3; the variations after
17.Qxb7 are quite pretty. 17.Qe3!
(17.Qxb7 Bxd4! 18.Qxa8+ Ke7 19.Qd5
Qb6 20.Rd1 (20.Qb5 Qc7 21.Qd5 Qb6©)
20...Bc5©)
17...0-0-0
(17...0-0 18.Kf2 Rae8 19.Bxe5 Rxe5
20.Qd4²)
18.Qc3!

19.Kf2 Rhe8 20.Bxe5 Rxe5 21.Qd4

Rde8©;
17.Bc3 Qc5 18.Bxe5 dxe5 19.b3 0-0
20.Qe3 Qa3! 21.Kf2 Rae8

This Qc3 move would not be available to

White were his pawn on b3, as it is in the (main)
variation 16...Be6 b3 17.Bf5.
(18.Kf2 Rhe8 19.Qc3! Qxc3 20.bxc3
(20.Bxc3 Bxc3 21.bxc3 Re5=) 20...Re7 21.Rd1
Kc7 22.Rd2 Rde8 23.g3 h5 24.Be2²)
18...Qxc3 19.bxc3 Kb8 20.Kf2 Rc8 21.h4² 22.Be2
17.Qe3 (22.Qxe5? Bf7 23.Qc3 Qc5+ 24.Kg3 Re3
25.Qb2 Rd8 26.Be2 Qg5+ 27.Kf2 Rxe2+–+)
17.Rd1 Bxc4+ 18.Kf2 Bf7 19.Bxe5 dxe5 22...e4!
20.Qxb7 0-0 21.Qb5 Qxa2 22.Qxe5 Rfe8 23.Qf6
Rab8 24.Rd2 Qa5=;
17.b3 Bf5! Unlike in the 16...Bf5 lines,
White will not have Qe3-c3 now. 18.Qe3
(18.Qxb7?? Bxd4 19.Qxa8+ Ke7 20.Qb7+
Kf6–+)
18...0-0-0

95

23.Rhe1
(23.Qxe4 Qc5+ 24.Kf1 Bh3 25.Qd3 Qe5
26.Qd5+ (26.Re1 Rd8 27.Qc2 Qe3 28.gxh3
Rd2–+) 26...Qxd5 27.cxd5 Bxg2+ 28.Kxg2 Rxe2+
29.Kg3 Re3 30.Rhf1 Rd3 31.Rad1 Rfxf3+
32.Rxf3 Rxd1=)
23...exf3 24.Bxf3 Bxc4 25.Qxe8 Rxe8
26.Rxe8+ Kg7 27.Ree1 Bf7
This position may not look equal; White
does after all have a slight material advantage, but
his rooks will be permanently tied down to the
defense of the queenside pawns and therefore
never able to get activated. The position is just a
draw. 28.Kg1
(28.Bxb7 Qc5+ 29.Re3 Qc7 30.Bf3
Qxh2=)
28...b5 29.Kh1 Qb2 30.Bc6 b4 31.Rf1 a5
32.Rae1 Bg8 33.Re7+ Kh6 34.Re8 Kg7=;
17.Bxe5 dxe5 18.Kf2 0-0-0 19.Qe3 Rd2+
20.Be2 Rhd8 21.Rad1 Qb6 22.Qxb6 axb6=
17...Bxc4+ 18.Kf2 Bxf1 19.Rhxf1
96
19.Bc3 Qb6 20.Qxb6 axb6 21.Bxe5 dxe5
22.Rhxf1 Kf7 23.Rfe1 Ke6 24.a3 Rhc8 25.Re2
Ra5 26.Rae1 Rc4=
19...0-0 20.Bxe5 dxe5
20...Qxe5 21.Qxe5 dxe5 22.Rfe1 Rae8
23.Re2²
21.Rfd1
The simplified nature of the position
should not induce a comatose attitude. From
Shereshevsky’s marveous book “Endgame
Strategy” we are shown the plight of the passive
defender; Black must be vigilant to avoid ending
up on the wrong end of a masterpiece.
21...Rac8!
In such endgames it is usually easiest to get
a draw by defending actively — Black threatens to
invade the 2nd rank, allowing White to invade the
7th.
21...Rad8 22.a3 Rxd1 23.Rxd1 Qa4 24.Rc1
Qh4+ 25.Kg1 Qf4 26.Qc3 Rd8 27.g3²;
21...Rae8 22.Qb3+
(22.Kg1 Rf7 23.a3 Re6 24.Kh1 Qb6
25.Qe2 Rfe7=)
22...Kh8 23.Qxb7
(23.Rd5 Qc7 24.Qe3 e4 25.Qd4+ Qg7
(25...Kg8 26.Rd7 Qe5 27.Qxe5 Rxe5 28.Rxb7²)
26.Qxg7+ Kxg7 27.Rd7+ Rf7 28.Rxf7+ Kxf7
29.f4 Ke6 30.Rd1 Rc8 31.Ke3 h5 (31...Rc2
32.Rd2 Rxd2 33.Kxd2±) 32.Rd2 Kf5 33.h3 Rc1
34.Rd5+ Ke6 35.Re5+ Kf6 36.Rb5 b6 37.Kxe4²)
23...Qc5+ 24.Kf1 e4 25.Rd7 Qh5 26.Qxa7
exf3 27.Qd4+ Re5 28.Re1 fxg2+ 29.Kg1 Rf1+
30.Rxf1 gxf1=Q+ 31.Kxf1 Qxh2 32.Re7 Qh1+
33.Kf2 Qh2+ 34.Kf1=
22.Rd7 Rf7 23.Rad1

23.Qb3 Qb6+ 24.Qxb6 axb6 25.Rxf7 Kxf7

26.Re1 Ke6 27.Re2 Rc4=
23...Rc2+ 24.Kg3

24.R7d2 Rxd2+ 25.Rxd2 e4!=

24...Qb5 25.R1d2 Rxd2 26.Rxd2

26...e4!! 27.Rd8+

27.f4 Qa5 28.a3 Re7=

27...Rf8 28.Rxf8+ Kxf8 29.Qxe4 Qxb2=

97
 6

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Nb5 d6!N 11.Qd2

I dub this the main line because it was

played in the only two tournament games of mine
in this variation.
11...Nxb5 12.cxb5 Be6!

14.exd5 Qxd5 15.Qxd5 Bxd5 16.0-0 e4!

17.Rfd1 Be6

White must already, as Carlsen often says,

“pull the emergency brake” so as to not end up
worse.

We continue to delay castling —

Lombardy would be proud. The importance of the
saved tempo is already felt: if 12...0-0 then we
transpose into the 10...0-0 line and White is
allowed 13.Bc4+=. It amazes me how good
Black’s position is here: it is already ‘zeroes’
according to the computer, and if White makes the
slightest inaccuracy (as both my strong opponents
did) then Black is already better. Once again, this 18.Bd4
opening forces us to look past our intuitive,
visceral response to each position, which might This was Nemcova’s choice.
mislead us into thinking White is better here due to 18.Rd2 Gorovets went with this one.
Black’s backward d-pawn and apparent lack of 18...0-0 19.a4 a6
counterplay. (19...f5 20.g3 (20.Bf4 Rfd8³) 20...Be5
12...0-0?! 13.Bc4² 21.Rc1 Rfd8 22.Rxd8+ Rxd8 23.a5 b6 24.b4²)
13.Be2 d5 20.bxa6 bxa6 21.a5 Rfb8!

98
 This nuance was preparation; I doubt I 23.Bd1
would have come up with the subtlety of luring
White’s bishop to b6 over the board. 22.Bb6 Rc8! 23.fxe4 Rdd2 24.Kf1 Bg4 25.Bxg4 Rf2+
Black threatens ...Bg7-h6. 23.Rad1 Bf6 Preventing 26.Kg1 Rxg2+ 27.Kh1 Rxh2+ 28.Kg1 Rcg2+
White from trading rooks with Rd8+, and 29.Kf1 Rxg4–+
preparing ...Be6-c4. 23...Rb2 24.Kf1?!
(23...Bh6? 24.Rd8+ Rxd8 25.Rxd8+ Rxd8
26.Bxd8±) 24.fxe4 Rdd2 25.Bf3 Rxa2µ
24.b4?! Bb3! 25.Rb1 24...Rdd2 25.Re2 e3!
(25.Bg4 Rc3 26.Rb1 Bg5„)
25...Bc3 White’s last chance to achieve
equality. 26.Rxb3?
(26.Rd7 Bc2 27.Rf1 Bxb4 28.f3 exf3
29.Bxf3 Bf5 30.Rd4 Bc3 31.Bxa8 Bxd4+ 32.Bxd4
Rxa8 33.Bf6=)
26...Bxd2 27.g4 Bg5! 28.b5 axb5 29.Bxb5
Bd8 30.Bd7 Rcb8 31.Bc6 Ra6–+ 0–1 (39)
Gorovets,A-Panjwani ,R Greensboro 2016.
18...0-0

This was as far as I had prepared (though it

was still time consuming for me to remember my
preparation over the board!).
19.Bxg7 Kxg7 20.f3?!
26.Rxd2 Rxd2 27.Be2 Kf6! 28.a4 Bxb3
20.Rd4 f5 21.Bc4 Bxc4 22.Rxc4= 29.Ra3 Rb2 30.g3 Rb1+ 31.Kg2 Ke5! 32.a5 Kd4
20...Rfc8! 21.b3 Rc2 22.Re1?!
0–1 Nemcova-Panjwani, Chicago 2014.
22.Kf1 Bf5³
22...Rd8!

99
 7
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Bd3
In my opinion this is White’s best move,
but authors like Khalifman recommend 10.Nb5
because they claim that “White should straighten
out the situation with the black knight on d4 at
once. Should it be left unmolested now, it will be
much harder to contend with it later.”
10...d6 11.0-0 0-0
White has many possibilities here, but
Black’s play is fairly straightforward regardless.
He will either play on the queenside with ...a7-a6
and ...b7-b5 or on the kingside with ...f7-f5 (and
often he combines play on both sides). However,
he should not rush with either of these plans,
because that would give White something to
respond to, while at present the onus is on White
to reveal how he plans to improve his position.
100
12.Qd2
12.Nb5?! Played by Almasi, Korneev, and
Izoria. With this move, however, White no longer
has an opening advantage. 12...Nxb5 13.cxb5 d5
14.exd5 Qxd5 15.f3 Be6 16.Qe2 Qd7 17.Rfd1
Qe7=;
12.Kh1?! Not a bad move in and of itself,
but it only makes sense in combination with f2-f4
(when the king is better placed off the g1–a7
diagonal) which yields White nothing here.
12...Be6 13.Qd2 Qd7
(13...f5?! Too soon. 14.exf5 Nxf5 15.Bxf5
gxf5 16.Bg5 Bf6 17.Bxf6 Qxf6 18.b3 Rad8
19.Rad1 a6 20.f4²)
14.f4 exf4 15.Bxf4
15...Nc6! This is the typical way to meet
White’s f2-f4. 16.Nd5 Bxd5 17.exd5 Ne5=;
12.f4?! exf4 13.Bxf4 Ne6
(13...Nc6!?=)
14.Be3 Be5 15.Qd2 Nc5 16.Bc2 Be6 17.b3
Qa5 18.Bd4 Rac8„;
12.Ne2 GM Shankland played this against
me in 2014. Our game continued... 12...Nc6
13.Qd2 Be6 14.b3 f5!
With White’s knight misplaced on e2 I felt
I could play ...f7-f5 earlier than usual. 15.Bg5?!
(15.f3 f4 16.Bf2 g5 17.Nc3 h5 18.Rfd1
Rf7„)
15...Qe8 16.exf5 gxf5 17.Bc2
(17.f3 Qh5 18.Be3 f4 19.Bf2 Rf6 20.Kh1
Kh8 21.Bg1 Rg8„; 17.Nc3 f4 18.Be4 Qh5
19.Qxd6 Rfe8 20.Bd5 Bxd5 21.Qxd5+ Kh8
22.Ne4 Rf8 23.Qd7 Rf7 24.Qh3 Qxh3 25.gxh3
Nd4³)
17...f4!
18.Qxd6
(18.Rae1 Qh5 19.h4 (19.Qxd6 Bf7µ)
19...Qf7 20.Qxd6 Bf5 21.Bxf5 Qxf5³)
18...Bf5 19.Bxf5 Rxf5 20.Qd5+ Kh8
21.Bxf4
101
I was not accurate here. I should have
continued with 21...Rd8!
(21...Ne7 This was my choice, but White is
close to equal here thanks to a surprise resource
that I missed. 22.Qxb7 exf4 23.Rae1 Rb8 24.Qxa7
Ra8 25.Qb7 Rb8 26.Qa7 Instead of repeating here
and accepting a draw I made a terrible
hallucination. 26...Re5?? 27.Nxf4 My board vision
failed me as I had missed White’s last move in this
variation. 27...Rxe1 28.Rxe1 Nc6 29.Rxe8+ Rxe8
30.Qe3™+–)
22.Qe4 exf4 23.Qxf5 Bxa1 24.Rxa1 Qxe2
25.Qf6+ Kg8 26.Qg5+ Kf7 27.Qxf4+ Ke8µ;
12.a4 Be6 13.a5 f5 White has effectively
wasted two tempi with a4-a5 so Black can play
...f7-f5 earlier than usual. 14.exf5
(14.Bxd4 exd4 15.Nd5 f4 16.Be2 Bxd5
17.cxd5 h5„)
14...Nxf5 15.Bxf5 gxf5„;
12.Rc1
Witness the battle between two former
World Chess Champions (though Smyslov got his
title the year after this game was played).
Botvinnik masterfully outplayed Smyslov but let
him get away with a draw in the end.
12...Be6 13.b3 a6 14.Bb1 Rb8 15.Kh1 b5
16.cxb5 axb5 17.Qd3 b4 18.Nd5 Bxd5
19.exd5 Qa5 20.Qc4 Rb5 21.Bd2 Rfb8 22.Qc8+
Bf8 23.Qd7 R5b7 24.Qg4 f5 25.Qh3 Qxd5µ
1/2–1/2 (46) Smyslov,V-Botvinnik,M Moscow
1956.
12...Be6
It is worth dissecting this position a little,
because it is one where general understanding of
themes is more important than concrete knowledge
of moves. Black’s counterplay will take root in
either (or both) ...f7-f5 or ...b7-b5, but suppose for
a moment that Black ‘does nothing’ — what will
White do? There are basically two plans at
White’s disposal that Black must constantly be
prepared to meet: White will either aim for Nc3-e2
102
or go for f2-f4. Black generally has three ways of
meeting Nc3-e2: capture on e2, ignore it and play
...b7-b5, or retreat ...Nd4-c6. Which of these Black
plays will depend on the concrete timing of
White’s Nc3-e2; sometimes all three options are
viable for Black. In response to f2-f4, Black will
either (and not both this time) play ...f7-f5, or he
will capture on f4 and re-route his knight from d4
to e5 via c6. Of course, Black should not ‘do
nothing’; he should remain flexible so as to be able
to play either ...f7-f5 or ...b7-b5 depending on
what White does. Black’s a8 rook will either be
placed on c8 or b8. Black’s Queen usually goes to
d7, but as we will see it can also find a home on
e7,c7, or a5. An important disclaimer is in order
before we continue: computers favor White by
approximately +0.4 in most lines. However, when
one presses the computer to formulate a plan, it is
totally at a loss: while maintaining its +=
evaluation, it recommends shifting White’s pieces
back and forth aimlessly. So, my point is, do not
be discouraged by the engine’s prejudice — it has
similar biases against the King’s Indian and other
respectable openings as well.
13.Rad1
13.Rac1 a6
14.b3
a) 14.Rfd1 The following is just one way
for Black to handle this position; he can
alternatively play 14...Rc8 or 14...Qd7. 14...Rb8
15.f3 (15.b3 Qd7 16.f3 b5 17.Nd5 Bxd5 18.cxd5
Rfc8 19.Qa5 Qb7=) 15...b5 16.cxb5 axb5 17.Nd5
Bxd5 18.exd5 b4 19.Rc4 Qa5 20.Bxd4 exd4 21.a3
Qxd5=;
 b) 14.f4 exf4 (14...f5?!
Even though this is suboptimal here, it is
worth seeing the variations which arise from it to
get a feel for the ‘other’ way of meeting White’s
f2-f4. 15.fxe5?! (15.Nd5! b5 16.Bxd4 exd4
17.exf5 Bxf5 18.cxb5 axb5 19.a3²) 15...dxe5
16.Nd5 White’s knight is vulnerable on d5, which
it would not have been had White played 15.Nd5
without exchanging on e5. 16...b5 (16...fxe4!?
17.Bxe4 Qh4„) 17.exf5 (17.cxb5 fxe4–+)
17...bxc4 18.fxe6 cxd3 19.Rxf8+ Bxf8 20.Qf2
Nxe6³) 15.Bxf4 Qa5! (15...Nc6!?= This is
perfectly playable as well.) 16.Bxd6 Rfd8 17.b4
Qh5
18.c5 (18.Bf4 a5! (18...Bxc4?? 19.Bxc4
Nf3+ 20.Rxf3 Rxd2 21.Bxd2+–) 19.b5 Bxc4
20.Bxc4 Nf3+ 21.Rxf3 Rxd2 22.Bxd2 Qc5+–+)
18...Rxd6! 19.cxd6 Be5 20.h3 Bxd6©;
14...Rc8!
103
Black intends to meet White’s Nc3-e2 by
taking on e2 and defending the d6 weakness with
...Rc8-c6.
a) 14...f5?! Black should not commit to
this so early; it just enables White to clarify his
intentions. 15.exf5 gxf5 (15...Nxf5 16.Bxf5 Bxf5
(16...gxf5 17.Bg5 Qd7 18.Rfd1±) 17.Qd5+ Rf7
18.f3±) 16.Bxd4 exd4 17.Ne2 d5 18.c5±;
b) 14...Rb8!? 15.Ne2 (15.Bxd4 exd4
16.Ne2 b5! 17.cxb5 axb5 18.Qb4 Qb6 19.Rc2
Rfd8 20.Nf4
20...d5! 21.exd5 Bxd5 22.Nxd5 Rxd5=
1/2–1/2 (39) Foisor,S (2258)-Melekhina,A (2205)
St. Louis USA 2016) 15...Nc6! (15...b5?! 16.cxb5
(16.Nxd4 exd4 17.Bh6 Bxh6 18.Qxh6 Qa5
1/2–1/2 (18) Korneev,O (2615)-Georgiev,V
(2519) Elgoibar 1999 19.cxb5 axb5 20.Bb1±)
16...axb5 17.Nxd4 exd4 18.Bh6 Bxh6 19.Qxh6
Qa5 20.Bb1 d5 21.e5±) 16.Rfd1
 (16.a4 f5„; 16.f4 exf4 17.Nxf4 Qd7 18.h3 17...Nxe2+! Without knights on the board
Ne5=) 16...Qe7!? (16...f5!? 17.exf5 (17.f3 f4 Black’s d5-square is less of a weakness. 18.Bxe2
18.Bf2 g5 19.Bb1?! g4! 20.Qxd6? Qe8µ) 17...gxf5 Qe7 19.Bb6 Rd7 20.a4 f5 21.exf5 gxf5„;
18.f4 (18.f3 Qe8 19.Bb1 Rd8 20.Nc3 Qf7 21.Nd5 13.Nd5
Kh8„) 18...Qe8 19.Nc3 Rd8 20.Nd5²) 17.Bb1
(17.Bb6 Bf6 18.Qb2 Bg5 19.Rb1 f5 20.Nc3 Qg7„)
17...Rbd8 18.f3 f5 19.exf5 gxf5 20.Nc3 Qf7
21.Nd5² Anything can happen, but objectively I’d
say White has better chances.;
15.f3
(15.Ne2 Nxe2+ (15...Nc6!? 16.Rfd1 Qe7
17.Bb1 Rfd8 18.Bb6 Rd7 19.Nc3 Qf8 20.Be3
Nd4„) 16.Bxe2 Rc6!

This was GM Sevian’s choice against me

in August 2016. Please note that our game was
somehow incorrectly recorded in the database as
an entirely different game altogether — clearly
someone confused our scoresheets with another
game’s! Our game continued... 13...Bxd5
(13...f5 14.Bxd4 Bxd5 15.exd5 exd4
16.Rfe1 Re8 17.Re6²)
14.cxd5
An excellent square for the rook: Black (14.exd5 Qd7=)
hinders White’s b3-b4 due to the c4 weakness, and 14...f5 15.Bxd4
defends d6 while preventing a future Be3-b6. (15.f3 f4 16.Bf2 Qd7 17.Rac1 g5„)
17.Rfd1 f5 18.exf5 gxf5 19.Bg5 Qe8„) 15...fxe4! 16.Bxe4 exd4 17.g3
15...Qd7 16.Rfd1
(16.Ne2 b5 17.Nxd4 exd4 18.Bh6 Bxh6
19.Qxh6 bxc4 20.bxc4 Rc5=)
16...Rfd8 17.Ne2
104
 25.Bxh7±)
23.Rxc1 Qxb2 24.Rc2
(24.Rf1 Qc3 25.Rd1 Qc7 26.Qxh5 Be5
27.f4 Bf6 28.Re1±)
24...Qb1+ 25.Kg2 Qd1 26.Bf5™ 26...d3
27.Rc7 Rg8 28.Qh6 1–0 Sevian,S-Panjwani,R
Greensboro 2016;
13.f4 exf4 14.Bxf4 Qa5! 15.Bxd6
(15.Rad1 Nc6! 16.Bxd6 Rfd8 17.c5 Bd4+
18.Kh1 Bxc5 19.Bxc5 Qxc5³)
15...Rfd8

17...Qb6
(17...Qd7 18.Rae1 Rae8 19.Qd3 Re7 20.f4
Bf6 21.Re2 Rfe8 22.Rfe1 Kg7

16.b4 Two moves are equally good here:

16...Qb6 is simpler to remember but 16...Qh5
gives White more chances to go astray.
(16.c5 16...Be5!! 17.Bxe5 Qxc5 18.Bxd4
Qxd4+ 19.Rf2 Qxd3=; 16.e5 Nc6µ)
16...Qb6
I was concerned about not having (16...Qh5 17.Bf4 (17.c5?! Rxd6! 18.cxd6
counterplay were I to go into this variation but it Be5 19.h3 Bxd6µ White is naked on the dark
turns out that passive defence suffices for Black. squares.) 17...Ne2+! 18.Nxe2 Bxc4 19.Ng3 Rxd3
23.Kg2 Qg4 24.h3 Qd7 25.h4 Qg4=) 20.Nxh5 Bd4+ 21.Qf2 gxh5 22.Rfd1 Bxf2+
18.Bd3! Rac8 23.Kxf2 Rad8=)
(18...Rae8! 19.Rac1 Qd8! 20.Rfe1 Qd7=) 17.Bc5 Qc7 18.Kh1
19.Rac1 Rc5 20.h4 Kh8? I thought I (18.Rad1 b6 19.Bxd4 Bxd4+ 20.Kh1 Bxc4
calculated this accurately but I missed White’s 21.Bxc4 Qxc4 22.Nd5 Be5=)
‘only-move’ on move 26. 18...b6 19.Bxd4 Rxd4 20.Nd5
a) 20...Qc7! The queen’s assistance is
needed on the kingside. 21.h5 (21.Rfe1 Rxc1
22.Rxc1 Qf7=) 21...gxh5 22.Qg5 Qf7=;
b) 20...Rfc8?! 21.h5 Rxc1 22.Rxc1 Rxc1+
23.Qxc1±;
21.h5 gxh5
(21...Rxd5 22.hxg6 hxg6 23.Kg2!+–)
22.Qg5 Rxc1
(22...Qc7 23.Qxh5 Be5 24.Rxc5 dxc5
105
 but if one is happy with a draw and comfortable
defending the resulting endgame, then this may be
the best option. 15.Ne2!N Unlike when Black’s
queen is on c7, he cannot here capture on e2
because the d6-pawn will fall (a defender on d8
can be hit by Be3-b6).
a) 15.Nd5 Bxd5 16.cxd5 Rac8=;
b) 15.f4 exf4 (15...Rae8!? 16.fxe5 dxe5
17.Nd5 b5 18.Nf6+ Bxf6 19.Rxf6 bxc4 20.bxc4∞)
16.Bxf4 Nc6N (16...Be5 17.Bxe5 dxe5 18.Qg5
Qc7 19.Kh1 b5 20.Nd5 Bxd5 21.cxd5² 1–0 (48)
Bellia,F (2424)-Royset,P (2215) Rethymnon 2011)
17.Be2 Ne5 18.Nd5 Rae8„;
15...b5!
20...Qxc4! 21.Rad1 Bxd5 22.Bxc4 Rxd2
23.Bxd5 Rxd1 24.Rxd1 Rc8=
13...a6 14.b3

In addition to being a good solidifying

move, White is now ready for Nc3-e2 (14.Ne2
b5!). Black has several ways of playing this
position. I have recommended what I consider to
be the simplest as the main line (14...Rc8), but
both alternatives are playable as well.

a) 15...Rfd8?! 16.Nxd4 (16.Bg5 Nxe2+

17.Bxe2 f6 18.Be3 Qc6 19.f3 f5„) 16...exd4
17.Bh6 Bh8 18.Qg5²;
b) 15...Nc6?! 16.Bb1±;
16.Nxd4
(16.Bxd4 exd4 17.cxb5 axb5 18.Bb1 d5!
19.Nxd4 dxe4 20.Bxe4 Rad8

14...Rc8!

Just as in the variation where White plays

13.Rac1, Black intends to meet Nc3-e2 by taking
on e2 and defending d6 with ...Rc8-c6!
14...Qd7 White’s attempts against this
move which are found in the database do not pose
problems for Black, but if White plays the novelty
15.Ne2!? then Black’s best option is to defend a
slightly worse but drawn rook endgame. I would
have felt bad recommending this as the main line,
106
 21.Nxe6 (21.Nf3 Qxd2 22.Rxd2 Rxd2
23.Nxd2 Rd8 24.Nf3 b4=) 21...Qxd2 22.Rxd2
Rxd2 23.Nxf8 b4 This endgame is easy for Black
to draw. 24.Nxh7 Kxh7 25.g3 Rxa2 26.Bd5 f5
27.Bc4 Bd4 28.Kg2 Kg7 29.Kf3 Kf6 30.h3 Kg7
31.g4 fxg4+ 32.hxg4 Ra1 33.Rxa1 Bxa1 34.g5
Bc3 35.Ke4 1/2–1/2 Brattain-Panjwani,
Philadelphia 2014)
16...exd4 17.Bh6 Bxh6 18.Qxh6 bxc4
19.Bxc4 Bxc4 20.bxc4 Qa7 21.Qd2
either rook. 15.Bxd4
a) 15.Ne2 Both 15...Nc6 and 15...Nxe2+
are playable; it is a just a matter of taste. 15...Nc6
(15...Nxe2+ 16.Bxe2 Rfd8 17.Qb4 Qe7 18.Bb6
Rd7 19.Rd2 Rc8 20.Rfd1 Rc6 21.Ba5 f5 22.Bf3 f4
23.Rd3 Bf8 24.a4 Qf7 25.Qd2 h5 26.h3 b6 27.Bb4
a5 28.Ba3 Qf6= It is difficult to see how either
side will make progress.) 16.f4 (16.Bb1 Rad8
17.Nc3 f5 18.exf5 gxf5 19.f4 Qf7=) 16...exf4
17.Nxf4 Ne5 18.Be2 Rad8 19.h3 Qc6!

The aforementioned slightly better, but 20.Nd5 Bxd5 21.Qxd5 Qc7 22.Bf4 Rfe8
with some accuracy drawn rook endgame starts 23.Bg5 Rd7 24.Rf2 h5 25.Bf4 Qc5=;
here. Black’s first task is to decide where to place b) 15.f3 Rac8 16.Ne2 (16.Rc1 Qd7 17.Ne2
his rooks. I give some sample variations, but I b5=) 16...Nc6 (16...Nxe2+!? 17.Bxe2 Rfd8 18.a4
make no claim to ‘solve’ this endgame. 21...Rfc8! Qe7 19.Qb4 f5 20.Rd2 Rc6 21.Rfd1 f4 22.Bb6
(21...Rab8 22.Qxd4 Qxd4 23.Rxd4 Rb2 Rd7
24.Rxd6 Rxa2 25.g3 Rc8 26.Rc1 Rc5 27.Rc3 Ra4
28.Rd8+ Kg7 29.Rd7 Raxc4 30.Rf3+–)
22.Qxd4 Qxd4
(22...Qc5 23.Qxc5 Rxc5 24.Rxd6 Rxc4
25.Re1 Rc2 26.a3 Rc3 27.Ra1 (27.a4 Rc4 28.a5
Ra4 29.Rd5 Rb8=) 27...Rac8 28.f3 R8c6 29.Rxc6
Rxc6=)
23.Rxd4 Rab8 24.g3!
(24.f3 Rb2 25.a3 Ra2 26.Rb1 Rxa3 27.Rb6
Ra4 28.Rbxd6 Raxc4 29.Rxc4 Rxc4 30.Rxa6 h5=)
24...Rc6 25.Rfd1 a5 26.Rxd6 Rxc4 27.e5
Rb2 28.a3 Rc5 29.Rf6 Rb7 30.e6 fxe6 31.Rxe6
Rc3 32.a4 Ra3 33.Ra6 Rxa4 34.Rd5 Kg7
35.Raxa5 Rxa5 36.Rxa5 h5=;
14...Qc7!? This was Dzindzichashvili’s White only has four pieces capable of
choice against my compatriot GM Lesiege in attacking the d6-pawn, and Black has four
1993. With the queen on c7, Black can respond to potential defenders. If White triples on the d-file
Nc3-e2 with either ...Nd4xe2 or ...Nd4-c6, since and aims for c4-c5 then Black will be able to
the d6-pawn can be defended in one move by obtain counterplay against White’s weak pawn on
107
b3 (with ...Qe7-f7 for example). A sample
variation may continue as follows. 23.a5 g5 24.h3
h5 25.c5 dxc5 26.Bxc5 Qxc5+ 27.Qxc5 Rxc5
28.Rxd7 Bxd7 29.Rxd7 Rxa5 30.Rxb7 Bf8=)
17.Bb1 Rfd8 18.a4 Rd7 19.f4 (19.Nc3 Nd4„)
19...exf4 20.Nxf4 Qa5=;
15...exd4 16.Ne2 Bg4 17.f3 Bd7 18.Bb1
(18.a4 b5 19.cxb5 axb5 20.axb5 Rfb8=)
18...b5
19.cxb5 Bxb5 20.Bd3 Qb6 21.Rf2 Rfc8
22.g3 Ra7 23.Kg2 d5 24.exd5 Rd7 25.Rc1 Rxc1
26.Nxc1 Rxd5 27.Bxb5 axb5= 0–1 (62) Lesiege,A
(2485)-Dzindzichashvili,R (2535) New York
1993.
15.f3
15.Ne2 b5!
(15...Nxe2+!? 16.Bxe2 Rc6∞)
16.Bxd4
(16.cxb5 Nxb5„)
16...bxc4!? 17.Bxc4 Bxc4 18.Be3 Be6
108
19.Qxd6
(19.Nc3 Qd7 20.Na4 (20.Nd5 Qb7=)
20...Rc6 21.Rc1 Qb7 22.Rxc6 Qxc6 23.Qd3 d5
24.exd5 Bxd5 25.Rc1 Qa8 26.Nb6 Be4! 27.Qc4
Qb7=)
19...Qxd6 20.Rxd6 Rc2 21.Nc1 Ra8
22.Rfd1 Bf8 23.R6d2 Rac8 24.Nd3 f6
Black is at little risk here; his two bishops
and active rooks provide fully adequate
compensation for the pawn. 25.f3 a5 26.Kf2 h5
27.Rxc2 Rxc2+ 28.Rd2 Rc6 29.f4 exf4 30.Nxf4
Kf7=
15...Qd7 16.Rfe1
16.Ne2 Nxe2+
(16...b5 17.cxb5 Nxb5 18.Qa5 Qb7∞)
17.Bxe2 Rc6 18.a4 f5„
16...Rfd8
16...b5!? 17.cxb5 axb5 18.Bxd4 exd4
19.Nxb5 Rc5 20.Na3 Rc3∞
17.Ne2 Nxe2+
17...b5 18.cxb5 Nxb5 19.Qa5 Ra8∞
18.Bxe2 Rc6 19.Bg5 Rf8 20.a4 f5„
109
 Appendix

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Bd3 0-0 11.0-0 b6!?

14.Bb1 f4!
(14...Nxb5 15.axb5 fxe4 16.Qg4± 1/2–1/2
(36) Honfi,K-Zaitsev,A Budapest 1963)
15.Bxd4 exd4 16.Ra3 Qf6 17.f3 a6 18.Nc7
Rac8 19.Nd5 Bxd5 20.cxd5 Rc5 21.Bd3 a5=
12...Bb7

I was pleasantly surprised to come across 12...Re8!? 13.Rfe1 Bb7 14.Rad1 a6 15.b3
this very rare idea in a book on pawn structures by Bc6
Andy Soltis. Soviet (grand) masters of the 1950s
and 1960s like Simagin, Bagirov, Aronson, and
Veresov developed some truly original ideas in
this system. Even if their crafty maneuvers are
ultimately ‘too good to be true’, it is instructive to
play over their games to get a feel for this system.
Black plans to continue with ...Bc8-b7, ...Rf8-e8,
...Ra8-c8, followed by ...Bf8-c5/...Nd4-e6, or
...Re8-e6-d6! I envy them for playing chess in the
pre-computer era; the analysis of this variation
must have brought them such joy. Computers do
spoil the fun a little, but there is no clear refutation
of the idea either.
12.Qd2
This is another typical idea of this line:
12.Nd5?! Bb7 13.f4 exf4 14.Bxf4 d6 Black plans to leave the pawn on d7, so the bishop
15.Qd2 Ne6 16.Bh6 Bxh6 17.Qxh6 Bxd5 18.exd5 is well placed on c6 from where it defends (and is
Nc5 19.Rf3 f5!³ 0–1 (69) Byvshev,V-Simagin,V defended by) the d7 pawn and prepares ...b6-b5.
Moscow 1952; 16.f3 Re6!?
12.a4 Bb7 13.Nb5 (16...Ne6 17.Qf2 Rb8 18.Nd5 b5 19.Bb6
(13.a5 bxa5 14.Qa4 Bc6 15.Qxa5 Qxa5 Qc8 20.Ba7 Rb7 21.Be3 Qb8 22.Qh4 Bxd5
16.Rxa5 Rfb8=) 23.exd5 Nf4 24.Bxf4 exf4 25.Rxe8+ Qxe8
13...f5! 26.Qxf4²)
17.Bf1 f5 18.Ne2 fxe4 19.fxe4 Nxe2+?!
(19...Rd6!?

110
 Kh8 19.b4 Na6 20.a3 Rfd8 21.Nc3 Nc7 22.Qd6
Qxd6 23.Rxd6 Rac8 24.Rd3 Ne6 25.Bxe6 dxe6
26.Rxd8+ Rxd8 27.Nb5 1/2–1/2 (27)
Roizman,A-Veresov,G Minsk 1955)
16.Nc3 Bf8! 17.exf5 gxf5 18.a3 Bc5

What a triumph of the artistic approach to

chess that this was Black’s best try. 20.Nxd4 exd4
21.Bg5 Qf8 22.c5!! Incredibly, Black can equalize
against all other moves! 22...bxc5 23.Bc4+ Re6
(23...Kh8 24.e5±) 24.Bxe6+ dxe6 25.e5²)
20.Bxe2 Qc7 21.Bg4 Rf6 22.Bxd7 Bb7 19.Bxc5 bxc5„;
23.Bg5 Rff8 24.Qd6 Qc5+ 25.Qxc5 bxc5 26.Be7 13.f3 Qe7 14.Rfd1 f5 15.exf5 gxf5 16.Nd5
Rf4 27.Be6+ Kh8 28.Bd5 1–0 (28) Qd6!=;
Kholmov,R-Aronson,L Riga 1954. 13.b3 f5 14.exf5 Qh4!
13.Rad1

It is not clear how Black should handle this

position. Based on my research, Black should
delay ...Ra8-c8 for some time.
13.Ne2 Ne6 14.Rfd1 f5 15.f3

15.f3
(15.fxg6?? Nf3+ 16.gxf3 Bxf3–+)
15...gxf5 16.Nd5 Bxd5 17.cxd5 e4„
13...Ne6

13...Bc6!?
15...Rf7! Black’s idea is to exchange dark
squared bishops with ...Bg7-f8-c5. Computers tend
to make us pessimistic about such maneuvers, but
in this particular position science and aesthetics
coincide.
(15...f4 16.Bf2 Qe7 17.c5 Nxc5 18.Bc4+
111
 14.f4
(14.b3 Ne6 15.Bc2 Re8 16.b4 Rc8 17.Bb3
Nd4 18.f3 Bf8 19.Bxd4 exd4 20.Ne2 b5 21.Nxd4
bxc4 22.Bxc4 Ba4 23.Bb3 Bxb3 24.Nxb3 Qb6+=)
14...exf4 15.Bxf4 Ne6 16.Bd6
(16.Bh6 Bxh6 17.Qxh6 Qe7 18.Nd5 Bxd5
19.exd5 Ng7=)
16...Re8 17.Kh1 Nc5 18.Qf2 Kh8 19.Qxf7
Qg5 20.Bc2 Nb7 21.Nb5 Nc5=;
13...Rc8?! 14.b3 Bc6
15.f4
(15.Bb1 Ne6 16.a4 Bf6! 17.Nb5 Ra8
18.Nd6 (18.g3 Be7 19.Nd6 Qb8 20.Bc2 Nd4
21.Bxd4 Qxd6 22.Bc3 Qe6 23.a5 bxa5 24.Bxa5
Bc5 25.Bb4 Qe7 26.Bc3 Rfb8 27.Ra1 Rb7=)
18...Bg5 19.g3 Bxe3 20.Qxe3 Qe7 21.f4 f6=)
15...exf4 16.Bxf4 Ne6 17.Bd6 Re8
112
18.Nb5 This is one reason Black’s rook
would have been better placed on a8.
(18.e5 Nc5 19.Qf4 Nb7 20.Nb5 Bxb5
21.Qxf7+ Kh8 22.cxb5 Nxd6 23.exd6 Qh4 24.g3
Qd4+ 25.Qf2 Qxf2+ 26.Kxf2 Re6 27.Bc4 Rf8+
28.Kg2 Rxf1 29.Bxf1 Bf8=)
18...a6 19.Na7 Ra8 20.Nxc6 dxc6 21.e5²;
13...f5
14.exf5 Qh4 15.Nd5
(15.fxg6?? Nf3+ 16.gxf3 Bxf3–+)
15...Nxf5
(15...gxf5 16.f4±)
16.Bxf5 gxf5 17.f4 e4 18.Bd4²
14.Bb1
14.b3 Bc6 15.f3 Re8
(15...Qe7 16.Bc2 Rfd8 17.Rfe1 Qa3!=
Black can ‘premove’ ...Bg7-f8.)
16.Nd5 Bf8 17.b4 Bg7 18.Rc1 Rc8 19.a4
Nd4 20.Rc3 a5 21.b5 Bxd5 22.cxd5 Bf8=
14...Bc6 15.b4 Rc8 16.a3
 16...Kh8!

16...Rc7 17.Ba2 Kh8 18.c5 Nd4 19.f4 bxc5

20.bxc5 Rb7 21.fxe5
(21.Nd5²)
21...Ne6 22.Bxe6 fxe6 23.Bg5 Qa5 24.Qd6
Rbb8 25.Be7 Rxf1+ 26.Rxf1 Re8 27.Rf3 Qxa3µ
1/2–1/2 (36) Smyslov,V-Bagirov,V Leningrad
1960.
17.Ba2

17.c5 bxc5 18.bxc5 Qa5„;

17.b5 Ba8 18.Qxd7 Qxd7 19.Rxd7 Rxc4
20.Nd5 Ra4=
17...f5 18.exf5 gxf5 19.f3 Qe8 20.Nd5

20.b5 Ba8 21.Qxd7 e4„

20...f4 21.Bf2

21...e4! 22.fxe4 Qh5 23.Rde1 Be5„

113
 CHAPTER 5

MAROCZY BIND: MAIN LINE

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0

This variation does not have an accepted

name, but because it is the top choice of virtually
all the top Accelerated Dragon players nowadays,
I will refer to it as the ‘main line’. Compared with
the Breyer Variation, this line is far more nuanced
and, frankly, difficult to play well (for both sides).
As with all practical matters, the choice involves a
tradeoff: the Breyer is easy to play for Black, but if
White plays accurately (which very few are able to
do in practice) then he can achieve a slightly more
pleasant position (if you find this state of affairs to
be unacceptable, I recommend abandoning the
Sicilian altogether in favor of the Berlin or
Contents Marshall). In this line Black’s counterplay is more
reactive in nature — Black can’t just ‘wear a
blindfold’ and proceed with a plan on autopilot,
1. 9...Nxd4 but in return Black can count on dynamic equality
2. 9...Bd7 10.sidelines and a tense struggle where all three results are on
3. 9...Bd7 10.Nc2 the table.
4. 9...Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 9.0-0
12.sidelines
5. 9...Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3 There are many ways to play this position
Nd7 13.sidelines for Black, but the main tabiya involves Black
6. 9...Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3 playing (in various move orders) ...Bc8-d7,
Nd7 13.Be3 ...Nc6xd4, ...Bd7-c6, ...a7-a5, ...Nf6-d7-c5. The
question of move order is an important one for
Black here: should Black start with 9...Bd7 or first
1 play 9...Nxd4 10.Bxd4 Bd7? In practice they
usually transpose, because most White players are
unwilling to play a different line against each of
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7 these, preferring instead to enter the Qd2 main line
5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 regardless of Black’s move order. That being said,
according to my research, 9...Bd7 is the
significantly better move order (and it is
introducted in the next chapter), in light of
114
9...Nxd4 10.Bxd4 Bd7 11.Qd3!.
9...Nxd4 10.Bxd4 Bd7

11...Bc6?! 12.b4!²;
11...a5 12.b3!
(12.a3 a4 13.c5 dxc5 14.Bxc5 Bc6 15.Qe3
11.Qd3! Qc8 16.f3 (16.Bxe7 Re8 17.Bxf6 Bxf6 18.f3
Qe6©) 16...Qe6 17.Rfd1 Rfd8=)
This is not a new move, it was first played 12...Bc6 13.a3 Nd7
in Smyslov-Pirc 1956 (1–0); for some reason it has (13...b6 14.b4 Nd7 15.Bxg7 Kxg7
not caught on. I owe Kevin Spraggett for revealing 16.Qd4+ Kg8 17.f4²)
its strength to me. White’s basic idea is to 14.Bxg7 Kxg7 15.b4 axb4
prophylactically defend the e4-pawn so that (15...b6²)
...Bd7-c6 can be met with b2-b4! White usually 16.axb4 Qb6 17.b5
wants to wait for Black’s bishop to be on c6 before
playing b2-b4, because the weak c4-pawn is under
less pressure when Black’s bishop is on c6 as
opposed to e6. In general Black wants to prevent
White from playing b2-b4 (at least) until his
knight has landed on c5, so the central theme of
this variation is White trying to achieve b2-b4 and
Black trying to either prevent it or render it
premature. Unfortunately, nothing seems to work
for Black. I will present the following lines
without comment, because I am recommending we
avoid this altogether, but I encourage the reader to
try to find improvements on my analysis, and to at
least skim through the ensuing variations to get a
feel for which positions are playable for Black and 17...Rxa1
which are to be avoided. (17...Ne5 18.Qd2 Bd7 19.Na4 Qa7 20.c5
11.b4 Be6 12.f3 Nd7 13.Nd5 Bxd4+ dxc5 21.Qd5 Qb8 22.Nxc5±)
14.Qxd4 Bxd5 15.cxd5 a5 16.a3 axb4 17.axb4 18.Rxa1 Ne5 19.Qd2
Rxa1 18.Rxa1 Qc7= (19.Qc2 Bd7 20.Nd5 Qd4 21.Rd1 Qc5
11...a6!? 22.Nxe7 Re8 23.Nd5 Bxb5=)
19...Bd7 20.Na4!
(20.Nd5 Qc5 21.Qe3 (21.Nxe7 Nxc4
22.Qc3+ Ne5 23.Qxc5 dxc5 24.f4 Bg4=) 21...Rc8
22.Qc3 (22.Nxe7 Nxc4=) 22...e6 23.Ne3 f6
115
24.Qd2 Be8 25.Rd1 Nf7 26.Bg4 Bd7=)
20...Qc7
(20...Qa7 21.c5 dxc5 22.f4 c4+ 23.b6±)
21.f4
(21.Qd4 g5 22.h4 h6=)
21...Ng4
22.Rd1!
(22.Qd4+ e5 23.Qb6 Qxb6+ 24.Nxb6 Nf6
25.Rd1 Be6 26.Bf3 Rb8! (26...exf4 27.Rxd6 Rb8
28.e5 Ne8 29.Rd4 g5 30.Nd7 Rd8 31.Bxb7 Rxd7
32.Rxd7 Bxd7 33.b6 Kf8 34.Bd5±) 27.Rxd6 Ne8
28.Rd1 exf4 29.e5 (29.c5 g5 30.h3 Nc7 31.Rb1
Ne8∞) 29...f6 30.exf6+ Nxf6 31.h4 h6 32.Rd6 Kf7
33.Nd5 Bxd5 34.Bxd5+ Nxd5 35.cxd5 Rc8
36.Rd7+ Kf6 37.Rxb7 Rd8 38.Rb6+ Kf5 39.d6 g5
40.hxg5 hxg5 41.Rc6 Ke5 42.b6 Rxd6 43.Rxd6
Kxd6 44.Kf2 g4 45.b7 Kc7 46.Ke2 Kxb7 47.Kd3
Kc6 48.Ke4 f3 49.gxf3 gxf3 50.Kxf3 This was not
a game, I just wanted to analyze at least one
variation ‘down to kings’ in this book.)
22...Nf6 23.Qd4 Bc8 24.e5
(24.Nc3 e5=)
24...dxe5 25.fxe5 Nd7 26.e6+ Nf6 27.exf7
e5 28.Qc3
(28.Qc5²)
28...Be6 29.c5 Nd5 30.Qb2 Bxf7 31.Bf3±
12.b4!
12.a4 a5 13.Nd5
(13.Rfd1 Bc6 14.b3 Nd7 15.Rab1 Nc5
16.Qe3 Bxd4 17.Rxd4 e5 18.Rd2 Qe7=)
13...Nxd5 14.exd5 Bf5 15.Qe3 Bxd4
16.Qxd4 Ra6=;
12.Qe3 Qb8!?
(12...b5 13.cxb5 axb5 14.Bxb5 (14.Nxb5
116
Ra4!©) 14...Bxb5 15.Nxb5 e5 16.Ba7 d5 17.Bc5
d4 18.Qd3 Re8 19.b4±)
13.Rfd1
(13.a4 a5 14.Rfd1 Bc6 15.Nd5 Re8!
(15...Bxd5 16.cxd5 Nd7 17.Bxg7 Kxg7 18.Bb5
Nc5 19.Rac1 f6 20.f4²) 16.Nb6 (16.b3 Nd7
17.Bxg7 Kxg7=) 16...e5 17.Bc3 Nxe4 18.Nxa8
Nxc3! (18...Qxa8?! 19.Be1 Nc5 20.b4 (20.Rxd6
Ne6©) 20...axb4 21.Bxb4 Bf8 22.a5±) 19.bxc3
Qxa8 20.Rxd6 Bf8 21.Rdd1 e4©)
13...Bc6
14.Rac1
(14.a4 a5 15.Nd5 Re8 (15...Bxd5 16.cxd5
Nd7 17.Bxg7 Kxg7 18.Bb5 Nc5 19.Rac1 Qd8
20.f4 f6 21.Rc3 Rc8 22.Rdc1 Rg8! 23.R1c2 b6
24.b3 Qf8! 25.Rc4 Kh8 26.g3 g5=; 15...e5 16.Bb6
Bxd5 17.cxd5 Nd7 18.Rac1 Nxb6 19.Qxb6 Bf6
20.g3 Bd8 21.Qb5 Qa7 22.Qd7 Qb8 23.Rd3 Bb6
24.Rb3 Rd8 25.Qb5 Bd4 26.Qxb7 Qxb7 27.Rxb7
Rdb8 28.Rxb8+ Rxb8=) 16.Nb6 e5 17.Bc3 Nxe4
18.Nxa8 Nxc3 19.Qxc3 Qxa8 20.Rxd6 e4©)
14...b5
(14...e5 15.Bb6 Nd7 16.Nd5 Bxd5
17.Rxd5±)
15.Nd5 Bxd5
(15...Re8 16.cxb5 Bxd5 17.exd5 axb5
18.Qb3 b4 19.Rc4± 1–0 (31) Gurevich,I
(2475)-Taylor Chicago 1992)
16.cxd5 Nd7 17.Bxg7 Kxg7 18.Qd4+ Kg8
19.Bg4 Nf6 20.Bh3 Qa7 21.Qd3 a5 22.g3 Rfb8∞;
12.f3
 13.a4 b6 14.Rfd1 Bc6 15.b4 a5 16.b5
Bb7=;
12.Rfd1 b5 13.cxb5 axb5 14.Nxb5 Bxb5
15.Qxb5 Nxe4 16.Bxg7 Kxg7 17.Bf3 Rb8 18.Qd3
Nc5=
12...Be6

12...b5 13.cxb5 axb5 14.Nxb5 Bxb5

15.Qxb5 Nxe4 16.Bxg7 Kxg7 17.Qd5 Nf6
18.Qb3±;
12...e5 13.Be3 Be6 14.Rac1 Rc8 15.Na4
b5 16.cxb5 axb5 17.Nc3 Bc4 18.Qd2±
13.Rfc1 Rc8 14.Nd5 Nxd5

12...Rb8
(12...b5?! 13.cxb5 axb5 14.Nxb5 Ra4
15.Nc3 (15.Na3 Qb8!©; 15.Be3 Bxb5 16.Qxb5
Qa8=) 15...Rb4 16.Rab1 (16.b3 Nxe4 17.Bxg7
Nc5 18.Qe3 Kxg7 19.Nd5 Rb8= Instinctively,
White appears to be better because of the
connected passed pawns; however, those pawns
are easily blocked and the dark squared
weaknesses in White’s position shouldn’t be
discounted.) 16...Qb8 (16...Nxe4 17.Bxg7 Nc5
18.Qe3 Kxg7 19.Nd5 Rb8 20.b4 Na4 21.b5 Nc5
22.Rfd1 Be6 23.Bc4 f6 24.Qd4 Bxd5 25.Bxd5
Qb6∞) 17.b3

15.exd5

15.cxd5 Bxd4 16.Qxd4 Bd7=

15...Bf5 16.Qd2 Bxd4 17.Qxd4 Re8

17...e5 18.dxe6 Bxe6 19.a3²

18.Bf3 a5 19.a3 b6 20.h4 h5 21.Re1²

17...Be6 (17...Rxd4 18.Qxd4 Ng4 19.Qd3

(19.Qd2 Qa7+ 20.Kh1 Nf2+ 21.Rxf2 Qxf2 22.Nd5
Bb5=) 19...Qa7+ 20.Kh1 Nf2+ 21.Rxf2 Qxf2
22.Nd5 e6 23.Ne3 Bb5 24.Qxb5 Qxe3 25.Qd3 Qf2
26.Rf1 Qa7 27.a4 d5 28.exd5 Rd8 29.Qa6 Qxa6
30.Bxa6 Rxd5 31.Bc4 Rd2 32.a5±) 18.Be3 d5
19.Bc5 dxe4 20.fxe4 Rd8 21.Qe3 Rb7 22.Rbd1
Rbd7 23.Kh1²)
117
 2 Nowadays I prefer to play 11...Bc6 here,
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0
Bd7
The downside of this line is 10.Nc2,
whereas the downside of the 9...Nxd4 move order
is 11.Qd3! In my judgment we should prefer to
allow White 10.Nc2 here. Let’s first have a look at
some side moves.
10.Nb3
The logic of this move is similar to 10.Nc2:
since White has more space, he wants to avoid
exchanges to keep Black’s position cramped.
Furthermore, the ‘soft’ c4-pawn can be defended
by Nb3-d2, as it often is by Nc2-a3 in the 10.Nc2
line.
10.Rc1 Nxd4 11.Bxd4
118
because things are likely to transpose to the
10.Qd2 line where White’s rook on c1 does
nothing to promote the central positional aim of
b2-b4. The reader may prefer 11...Bh6!? which I
have played a few times as well. The choice is just
a matter of taste. 11...Bh6!?
(11...Bc6 If Black plays this then the
10.Rc1 line has little independent value. 12.f3 a5
13.b3 Nd7 14.Be3 Nc5 15.Qd2= This is a
transposition to the 10.Qd2 line, where White’s
a-rook does not tend to go to c1.)
12.f4
(12.Rc2 e5 13.Be3 Bxe3 14.fxe3 Bc6
15.Qd3 (15.Bd3 a5 16.Rcf2 Nd7 17.Qg4
Nc5 18.Bc2 Bd7 19.Qd1 Be6 20.Nd5 Kg7 21.b3
Bxd5 22.Qxd5 f6 23.a3 Qb6 24.Rb1 Qc6=) 15...a5
16.Rd2 Ne8 17.Bd1 Qb6 18.Rdf2 Rd8 19.Bg4 Nc7
20.h4 Na6 21.h5 Nc5 22.Qc2 Qb4 23.Nd5 Bxd5
24.exd5 e4 25.Be2 (25.Rf6 Nd3 26.b3 a4„)
25...a4∞)
12...Bc6 13.Bxf6
a) 13.e5 dxe5 14.Bxe5 Nd7 15.Bd4 e5!?
(15...Re8!? 16.Rc2 e5 17.fxe5 Nxe5„) 16.fxe5
Qg5 17.Nd5 Bxd5 18.cxd5 Qxc1 19.Qxc1 Bxc1
20.Rxc1 Nxe5 21.Rc7 (21.d6 f6µ) 21...f6 22.Rxb7
(22.d6 Rac8 23.Rxb7 Rf7 24.Rxa7 Rxa7 25.Bxa7
Rc2µ) 22...Rf7 23.Rb5³;
b) 13.Qd3 e5 14.Be3 exf4 15.Bxf4 Bxf4
16.Rxf4 Nd7=;
13...exf6
 14.Qd4 This was American prodigy
Ruifeng Li’s choice against me.
a) 14.Qd2 Re8 15.Bf3 Qb6+ 16.Kh1 Rad8
17.g3 Bg7 18.Rc2 Qc5 19.Nd5 a5=;
b) 14.Kh1 We follow one of my games
here. 14...Re8 15.Bd3 Qb6 16.Rc2 a5 17.Rd2 a4
18.Bb1 Bf8 19.b3 axb3 20.axb3 Ra1 21.Nd5 Bxd5
22.Rxd5 Qe3? (22...Rxe4 23.Bxe4 Rxd1 24.Rdxd1
f5 25.Bd5 Bg7 26.b4 Bf6 27.b5 Kg7=) 23.Rf3
(¹23.c5!) 23...Qa7 24.Qc2 f5 25.Rf1 Rxe4
(25...fxe4 26.f5 e3 27.fxg6 hxg6 28.Rxf7
Rxb1+–+)
26.g4!! Qe3 27.Rd3 Qe2 (27...Rxf4
28.Rxe3 Rxf1+ 29.Kg2 Rfxb1∞) 28.Qxe2 Rxe2
29.gxf5 Rb2µ Shlionsky-Panjwani, New York
2016 (1/2–1/2);
14...Re8 15.Rcd1 Qa5 16.Bd3 Bg7 17.Qf2
a6 18.a3
119
18...f5! 19.exf5 Bxc3
(19...gxf5 20.Nd5 Bxd5 21.cxd5 Qxd5
22.Bxa6=)
20.bxc3 gxf5 21.Rde1
a) 21.Qg3+ Kh8 22.Qg5 Rg8 23.Qf6+ Rg7
24.g3 (24.Bxf5?? Bxg2–+) 24...Qxa3=;
b) 21.Qh4 Re6³;
21...Qxa3 22.Bxf5 Qxc3µ Ruifeng Li —
Raja Panjwani, Philadelphia 2014 (0–1);
10.f3 (As with the 11.Qd3 line which we
avoided by playing 9...Bd7 instead of 9...Nxd4,
White prophylactically defends the e4-pawn in
anticipation of ...Bd7-c6). 10...Nxd4 11.Bxd4 a5 If
White plays ordinary developing moves like
12.Qd2 here then we will transpose into material
we will discuss later, so we will just consider
attempts by White to obtain immediate play.
(11...Bc6?! 12.b4²)
12.c5
(12.b3 Bc6 13.a3 Nd7 14.Bxg7 Kxg7
15.Qd4+ Kg8 16.b4 Qb6 17.Qxb6 Nxb6=)
12...dxc5 13.Bxc5 Bc6
 14.Qb3
(14.Qxd8 Rfxd8 15.Bxe7 Rd2=; 14.Rc1
Nd7 15.Be3 a4 16.Qc2 Qa5=)
14...Nd7 15.Be3 Qb8! 16.Rad1 a4 17.Qc4
Rc8 18.Nd5 Bf8!
Black may optically appear to be passive
but his compactness is deceptive; all of Black’s
soft spots have been covered and White has no
targets. 19.Nb4 e6 20.Nxc6 Rxc6 21.Qb5 Bc5
22.Bxc5 Nxc5=
10...a5!
120
Black immediately targets White’s b3
knight and threatens to gain further space on the
queenside with ...a5-a4 and ...Qd8-a5.
11.Nd2
11.f3?! a4 12.Nd2?! a3 13.bxa3 Ng4
14.fxg4 Bxc3µ;
11.a3 a4 12.Nd2 Qa5= Ideas for Black
include ...Be6/...Nd7-c5 or alternatively
...e7-e5/...Nc6-d4.
11...Bc8!
Black’s bishop is misplaced on d7 in both
the 10.Nc2 and 10.Nb3 lines. Ordinarily Black
would prefer to play ...Be6/...Nd7-c5 but there is
no time for that here since White’s f4-f5 comes too
fast.
11...Be6 12.f4 Nd7?? 13.f5+–
12.Rc1 Nd7 13.Nb3
13.f3 Bd4!N 14.Bxd4 Nxd4 15.Ndb1 e5
16.Na3 Nc5 17.Nab5 Nce6=
13...a4 14.Nd4 Nxd4 15.Bxd4
 ...f7-f5 because that would undouble White’s
e-pawns, but White has nothing constructive to do
either. Black can literally play ...Ra6-a8-a6-a8 and
wait for White to do something.
22.Bxe6

22.Nd5? Bxd5 23.exd5 f5µ

22...fxe6 23.Rxf8+ Kxf8 24.Nb5 Kg7
25.Qf3 Rc6 26.Rf2 Rc8 27.Nxd6

27.g4 Rf8 28.Qxf8+ Qxf8 29.Rxf8 Kxf8

30.Nxd6 Ke7 31.Nb5 Nxe4³
27...Qxd6 28.Qf6+ Kg8 29.Qf7+ Kh8
30.Qf6+ Kg8=
15...Bh6!

This is a fairly common idea when White’s

rook is on c1. Black wants to play ...e7-e5 next,
when White has to accept doubled, isolated
e-pawns.
16.Rc2

16.f4 e5 17.Be3 exf4 18.Bxf4 Qb6+

19.Kh1 Bxf4 20.Rxf4 Qxb2 21.Rc2 Qb6 22.Rd2
Ne5 23.Nd5 Qc5 Black’s dark squares are in
theory weak, but without White’s dark squared
bishop on the board Black has nothing to worry
about. On the other hand, Black’s light squared
bishop is clearly superior to its counterpart.
24.Nf6+ Kg7 25.Rxd6 Be6=
16...e5 17.Be3 Bxe3 18.fxe3 Nc5 19.Rd2
Ra6 20.Bg4 Be6 21.Qe2 Qe7

It is hard to see how either side will make

progress; Black doesn’t particularly want to play
121
 3
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0
Bd7 10.Nc2
This move was a real headache for me to
deal with when preparing this book. I did not have
the luxury of recommending the usual 9...Nxd4
because I was in on the secret of Smyslov’s
powerful 11.Qd3! The database games were not
much help either; the consensus seems to be that
Black must sit back passively and demonstrate
resiliency in the face of White’s b2-b4 or
f4-f4/g2-g4 expansions (see the alternatives to
10...Rc8 for an illustration). I analyzed all kinds of
weird ideas for Black, like
...a7-a5/...Nc6-b4/...Bd7-c6/...Nf6-d7-c5, but I
realized that White’s position is too solid for Black
to get away with unprincipled play. I concluded
that if Black is going to be able to equalize in this
line, it will be via ‘healthy’, thematic moves in the
spirit of the opening.
10...Rc8!
122
Until around 2012 this was by far the most
popular move, and it still is overall, but it has been
absent from the top games in the last five years,
particularly in favor of 10...Qa5. I find the
positions after 10...Qa5 to be overly provocative:
Black does not need to cede as much ground as he
tends to in those lines. After 10...Rc8 we will
consider White’s three most popular attempts.
10...Qa5 11.Qd2 Rfc8 12.Rac1 Ne5 13.b3
Nc6 14.b4 Qd8 15.f4 Bg4 16.Bd3 Be6 17.h3 a5
18.a3 axb4 19.axb4 Nb8 20.Nd5 Bxd5 21.cxd5
Ra2 22.Qe1 Nbd7 23.Nd4 Ra3 24.Qd2 Rca8
25.Kh2 Ra2 26.Rc2 Rxc2 27.Qxc2 Qb6 28.Qc4
Qd8 29.Rc1± 0–1 (42) Swicarz,M-Cyborowski,L
Poland 2015;
10...a6 11.f3 Na7?! 12.Na3 Bc6 13.Qd2
Nd7 14.Rab1 Qb8?! 15.Rfc1 Rc8 16.b4 b6 17.c5
dxc5 18.bxc5 Nb5 19.Naxb5 axb5 20.cxb6 Nxb6
21.Nxb5 Be8 22.Rxc8 Nxc8 23.Qc1!+– 1–0 (47)
Vuckovic,B (2558)-Abramovic,B (2455) Kraljevo
SRB 2015.
11.f3
This is by far White’s most popular choice.
Against White’s other moves Black plays
...Nc6-e5, but here since White has committed
himself to f2-f3, Black can play 11...Be6! when
White’s best plan is to go for f4-f5, now at the cost
of a tempo.
11.Qd2 Ne5
 16...dxe5 17.Bxa7 Qc7 18.Bf2 Be6!
19.Nb5 Qb7 20.Nd6 Nxe4!³;
11.Rc1 Ne5

12.b3 White’s most popular move.

a) 12.Na3 Be6 13.b3 Qa5 14.Nab5 a6
15.Nd4 b5=;
b) 12.c5 Bc6! (12...dxc5 13.f4 Nc6 (11...a6!? 12.f4 b5 13.cxb5 axb5 14.Bxb5
(13...Neg4 14.e5 Nh5 15.h3 Nxe3 16.Qxe3 Bh6 Ng4 15.a4 Nxe3 16.Nxe3 Nd4 17.Bxd7 Qxd7
17.Rad1 Qe8 18.Bxh5 gxh5 19.Qf3²) 14.Rad1 Nb8 18.f5 Qa7 19.Kh1 e6 20.Rb1 Rfd8©)
15.Qe1 (15.e5 Ng4„) 15...Qe8 16.Qh4 b5 12.b3 This line may be one reason that
(16...Bc6 17.f5 Nbd7 18.Ne1²) 17.f5 b4 18.Nd5 Black stopped playing 10...Rc8, but I have found a
Nxd5 19.Rxd5 Bxb2 20.e5∞) 13.cxd6 Bxe4 new move which seems to revive Black’s position.
14.Bxa7 Qxd6 15.Qxd6 exd6 16.Ne3 Bd3=; (12.Na3 Qa5 13.f3 Be6 14.Nd5 Nxd5
12...b5! 13.f4 15.exd5 Bd7
(13.cxb5? Qc7 14.Bd4 Nxe4 15.Nxe4
Qxc2 16.Qe3 f5!µ)
13...Neg4 14.Bd4
(14.Bxa7 bxc4 15.b4 Bc6³ 1/2–1/2 (38)
Nijboer,F (2525)-Bosch,J (2425) Amsterdam
1996)
14...bxc4 15.Bxc4

In general Black has three plans in such

positions: either play ...f7-f5-f4 securing the e5
square for Black’s knight, or play ...e7-e6 opening
up the center with White’s knight misplaced on a3,
and finally ...a7-a6/...b7-b5 as usual. Here we will
see Black implement a different idea from these,
which works because of the concrete features of
15...e5! 16.fxe5 the position. 16.h3 (16.f4 Ng4 17.Bxg4 Bxg4
(16.Bxa7? Qa5 17.Be3 Nxe3 18.Nxe3 exf4 18.Qxg4 Bxb2 19.Nb5 Bxc1 20.Rxc1 Rxc4!–+)
19.Rxf4 Nxe4–+) 16...g5!?
123

Not the only move for Black, but certainly
his most aesthetically appealing option. 17.Qe1
(17.Rc2 h6 18.f4 gxf4 19.Rxf4 f5„; 17.Bxg5?!
Qb6+ 18.Rf2 h6 19.Bd2 Qxb2 20.Nb1 Qxa2³)
17...Qxe1 18.Rfxe1 f5 19.Bxg5 (19.Bxa7 g4©)
19...Nf7 20.Bxe7 Rfe8 21.Bd1 Bxb2 22.Rb1 Bxa3
23.Rxb7 Ne5 24.Rxe5 Bc5+ 25.Kf1 dxe5 26.Bxc5
Rxc5 27.Rxd7 Rxc4 28.Rxa7 Rd4=)
12...Qa5!N
An important novelty. White is forced to
make a decision of how to deal with the hanging
c3-knight, and Black will respond accordingly.
a) 12...a6 13.Nd4! Were it not for this
move, 12...a6 would be fine for Black. 13...Qa5
(13...Neg4 14.Bd2 Qb6 15.Nc2²) 14.Nd5! Nxd5
(14...Qd8 15.f3 Nxd5 16.exd5 e6 17.dxe6 fxe6∞)
15.exd5 f5 16.a4 Ng4 17.Bxg4 fxg4 18.Re1²;
b) 12...b5?! This is thematic but
unfortunately Black’s compensation is inadequate.
13.Nxb5 Nxe4 (13...Bxb5 14.cxb5 Nxe4 15.f4
Nd7 16.Nd4±) 14.Nxa7 Ra8 15.a4 Nc5 16.Bxc5
124
dxc5 17.Nb5 Qb6 18.Qd2 Rfd8 19.Qe3 Bf5
20.Nb4 e6 21.Rfd1 Rab8 22.Nd3²;
13.Qd2
a) 13.Nd4 Neg4! (13...a6 14.Nd5! (14.h3
b5 15.f4 Nc6 16.cxb5 Nxd4 17.Bxd4 axb5=; 14.f3
b5 15.cxb5 axb5 16.Ndxb5 Bxb5 17.Nxb5 Rxc1
18.Qxc1 Qxa2=) 14...Rfe8 15.b4 (15.a4 Nxd5
16.exd5 e6 17.dxe6 (17.f4 Nxc4! 18.bxc4 exd5µ)
17...fxe6 18.f4 Nf7 19.Bf3 d5=) 15...Qd8 16.f4!
(16.Nxf6+ exf6! 17.Qd2 Nc6 18.f3 f5„) 16...Neg4
(16...Nc6 17.Nc2±) 17.Nxf6+ Nxf6 18.e5 Ne4
19.e6 fxe6 20.Bg4 Bxd4 21.Bxd4 Nf6 22.Be2 Rf8
23.Qb3²) 14.Bd2 Qc5 (14...Nxh2 15.Ncb5 Qb6
16.Kxh2 a6 17.Be3 axb5 18.Ne6 Qa5 19.Nxf8²)
15.Nf3 (15.Nc2 Compare this to the analogous
position in the 12...a6 line, where Black’s queen is
on b6 instead of c5. There, Black’s next move is
unavailable to him. 15...Qe5! 16.g3 Qh5 17.h4
Qc5=) 15...Rfe8 16.h3 Ne5 17.a4 a6 18.Nxe5
Qxe5 19.Bd3 Qh5! 20.Qxh5 Nxh5=
In general Black should be happy to
exchange queens in this line of the Maroczy. Here,
Black will consider relocating his h5 knight to c5,
or perhaps instead he will relocate his g7 bishop to
a7 and keep the knight on f6. Eventually things
will fizzle out: White will play Nd5, Black will
take it with either knight or bishop, and rooks will
get exchanged on the c-file. As always, Black’s
counterplay is to be found in ...f7-f5 or ...b7-b5.;
b) 13.Bd2 Qb6 14.Nd5 Nxd5 15.exd5 f5„;
13...b5!
 Now that White’s c3-knight is ‘pinned’,
Nc3xb5 is no longer possible as it was after
12...b5. 14.Nd5
(14.b4 Qc7 15.c5 Rfd8„; 14.Nxb5?! Qxd2
15.Bxd2 Nxe4 16.Be3 a6 17.Nbd4 f5„)
14...Qxd2 15.Bxd2
(15.Nxe7+ Kh8 16.Bxd2 Rce8 17.Nd5
Nxe4=)
15...Nxd5 16.cxd5 f5 17.exf5 gxf5 The
position is dynamically equal. Black has obvious
static weaknesses, but he also has more central
pawns and active pieces. I will just give a sample
variation for how play could continue.
18.Nd4 a6 19.a4 bxa4 20.bxa4 Bxa4
21.Bxa6 Rxc1 22.Rxc1 Bd7 23.Nc6 Ra8 24.Bb5
Kf8 25.Bg5 Ng6=
11...Be6!
11...a6
125
12.Rc1!
(12.Qd2 Ne5 13.Na3 (13.b3 b5 14.cxb5
Qa5 15.b4 Qc7 16.Bd4 Nc4! (16...Bxb5 17.Ne3
Nc4 18.Nxc4 Bxc4 19.Rac1 Qb7=) 17.Qc1 axb5„)
13...Be6 14.Nd5 Bxd5 (14...Nfd7?! 15.Rac1
(15.Bd4 Nc6 16.Bxg7 (16.Be3 f5 17.Nf4 Bf7∞)
16...Kxg7 17.Kh1 a5 18.Nc2 Nc5 19.Nce3 f6²)
15...f5 16.exf5 gxf5?! (16...Bxf5 17.Rfd1²) 17.f4
Nc6 (17...Ng4 18.Bxg4 fxg4 19.f5 Bf7 20.b3 Nf6
21.Nxf6+ exf6 22.Rf4 h5 23.Rd4±) 18.Bf3 Nc5
19.Rfe1 Bf7 20.b4 Ne4 21.Bxe4 fxe4 22.b5 Nb8
23.Bb6± 1–0 (53) Young,G-Panjwani,R Las
Vegas 2013) 15.exd5 (15.cxd5 e6=) 15...e6
16.dxe6 fxe6 17.Rad1 (17.b4 d5=) 17...d5 18.c5
Nfd7 19.Rc1 (19.b4 a5=) 19...d4 20.Bg5 (20.Bxd4
Nxc5³) 20...Nf6=)
12...Ne5?!
(12...Be6 A recommendation of
Khalifman. 13.Nd5! (13.Na3 Nd7 14.Qd2 Qa5
15.Rfd1 f5! 16.exf5 Bxf5 17.Qd5+?! Qxd5
18.cxd5 Nb4 19.g4?! Bd3 20.Bxd3 Rxf3µ)
13...Nd7 14.b4 f5 15.exf5 Bxf5 16.b5±)
13.c5!! A brilliant computer-move!
(13.Na3 Be6 14.Nd5 Nfd7 15.f4 Nc6=
Black will continue with ...f7-f5; White’s knight is
awkward on a3.)
13...dxc5 14.f4 Neg4
(14...Nc6 15.e5 Ne8 16.Bf3 b6 17.Rf2!
Be6 18.Qxd8 Nxd8 19.Na4 Rb8 20.b4±)
15.e5 Nxe3 16.Nxe3 Ne8 17.Rc2! b5
18.Rd2 Rc7 19.a4±;
11...Na5 12.b3 a6
(12...Ng4 13.Bd2! Qb6+ (13...Ne5 14.Rc1
(14.f4 Nexc4!„) 14...f5 15.f4 Nf7 16.exf5 Bxf5
17.Ne3±) 14.Kh1 Nf2+ 15.Rxf2 Qxf2 16.Nd5+–)
 13.Qd2 b5 14.cxb5

Prophylaxis! Black would like to play

14...Qc7 13...Qa5 instead, but that allows 14.b4! which
(14...Rxc3 15.Qxc3 Nd5 16.Qd2 Nxe3 works because of Black’s e7-pawn being weak.
17.Nxe3±; 14...Ng4 15.Bd4 e5 16.Ba7!±) (13...Qa5?! 14.b4! Nxb4 15.Nd5 Nc6
15.Bd4 Bxb5 16.Qxa5 Nxa5 17.Nxe7++–)
(15...axb5 16.Nb4 Qb8 17.Kh1 Nc6 14.b3
18.Nxc6 Bxc6 19.a4±) (14.Rfd1 Qa5=; 14.f4 Nc5 15.Nd4 Nxd4
16.Bxb5 axb5 16.Bxd4 Qb6 17.Be3 Qb4=)
(16...e5 17.Nb4 exd4 18.Ncd5 Nxd5 14...Qa5 15.Nd4 Nxd4 16.Bxd4 Bxd4+
19.Nxd5 Qa7 20.Bd3 Nc6 21.f4 Ne7 22.Nxe7+ 17.Qxd4 Qc5 18.Qxc5 Nxc5=
Qxe7 23.Bxa6 Rc3 24.Bd3±) 12...Qa5!
17.Rac1 Qb7 18.Nd5 Nxd5
(18...e5 19.Be3 Nxd5 20.exd5 Ra8
21.Nb4±)
19.Bxg7 Kxg7 20.exd5±;
11...Ne5 12.b3 a6 13.Nd5 Nxd5 14.exd5 f5
15.Qd2 b5 16.cxb5 axb5 17.Nd4²
12.f4

This is the computer’s recommendation —

machines don’t have egos about blatantly losing
tempos.
12.Qd2 Nd7 13.Rac1
(13.Rab1 f5 14.exf5 Bxf5=; 13.Nd5 Bxb2
14.Rab1 Bg7 15.Rxb7 Nc5=)
13...Re8!
Black threatens a discovery attack on the
c4-pawn with either ...Nc6-b4(b8), when White’s
usual b2-b3 is impossible because of the hanging
c3 knight.
12...a6 This followed by 13...b5 almost
works but not quite. The opposite colored bishop
endgame which results is highly unpleasant for
Black, even if objectively drawn. 13.Rc1
(13.Qd2 Na5 14.b3 b5 15.cxb5 axb5
126
16.Bxb5 Qc7 17.Bd4 Bxb3!µ) c4-pawn would be hanging.
13...b5 14.cxb5 axb5 15.Bxb5 Ng4 16.f5! 14.b3
(16.Bxc6 Nxe3 17.Nxe3 Qb6 18.Qd2
Rxc6©) 14.Na3 Nxe4! 15.Nxe4 Bxb2 16.Nb5 Bxc1
16...Nxe3 17.Nxe3 Qb6 18.fxe6 Qxe3+ 17.Qxc1 Qxa2 18.Nbc3 Qa5„
19.Kh1 fxe6 20.Rxf8+ Kxf8 21.Na4 Na7 14...Ng4 15.Bxg4 Bxg4
22.Rxc8+ Nxc8 23.Qf3+ Qxf3 24.gxf3 It won’t
surprise me if Black can find a plan to force a
draw in this endgame, but fortunately we don’t
need to in light of 12...Qa5! 24...Bd4
(24...d5!?∞)
25.Bd7 Nb6 26.Nxb6 Bxb6 27.Bxe6 Bd4
28.b4±
13.Rc1!

The most logical choice: White takes his

rook off the h8-a1 diagonal on which it was
vulnerable after b2-b4, and by X-ray defends both
the c3-knight and c4-pawn.
13.a3 Nb8!

16.Qd2!

16.Qxg4 Bxc3 17.a4 White is searching for

harmony, and Black must act fast to prevent White
from finding it. 17...Na5!
(17...Qa5 18.Rcd1 Bg7 19.Qe2²)
18.Rb1

14.Nd5
(14.b4 Qd8³ White’s c4-pawn falls.)
14...Qd8 15.Bd3
(15.f5 Bd7 16.Bxa7 Nxd5 17.Qxd5 Bc6
18.Qd3 Bxb2 19.Rab1 Be5=)
15...Nbd7 16.Nd4 Bg4=;
13.g4 Nb4!
(13...Nb8!?)
14.Nxb4 Qxb4 15.a3 Qa5 16.Nd5 Nxd5 18...Nxb3! 19.Rxb3 Qxc4 20.Rfb1 Qxe4
17.exd5 Bd7= 21.Qe2 Rc4 22.Na3 Rxa4 23.Rxc3 Rxa3 24.Rbc1
13...Qa6! Rxc3 25.Rxc3 a5=
16...Bd7 17.f5 Qa5 18.Nd4 Rfe8! 19.Rcd1
13...Nb4?! 14.Nxb4 Qxb4 15.b3²; a6 20.a4 Ne5
13...Nb8?! 14.Nd5 Qd8 15.Nxf6+ Bxf6
16.Nd4² Note that were White’s rook not on c1 the 20...Be5 21.Nd5 Qxd2 22.Rxd2 Bxd4
127
23.Bxd4 Nxd4 24.Rxd4 Bc6 25.Nb6 Rcd8
26.Rfd1
(26.b4 e5 27.fxe6 Rxe6=)
26...a5 27.c5²
21.h3 Bc6²

This is one of the few times in the book

where I’ll give White a slight advantage in a main
recommendation. The reader should be suspicious
of authors who claim pure equality in all lines in
anything other than the most topical lines of the
Ruy Lopez or perhaps the Najdorf Sicilian. This is
the sort of += we have to live with as Accelerated
Dragon players. White played very well to get
here; he had to navigate his way around many
landmines. After all that, we can seek consolation
in the fact that objectively speaking, there are only
three results in chess, and ‘+=’ is just short-hand
for ‘= after accurate play’. The onus is on White to
improve his position; Black can shift his pieces
around until White makes a concession. For
example, if White goes for g4-g5, then Black may
relocate his knight to c5 via d7, and place his
bishop on e5, perhaps combined with ...Qa5-b4.

128
 4 13.Rae1

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0
Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6

White is anticipating Black’s ...Nf6-d7, and

preemptively initiating his thematic rook lift.
Black, however, need not comply.
13.b3 Nd7 14.Bxg7 Kxg7

Here White’s main alternative to 12.f3 is

12.Bd3.
12.Bd3

A move first played by GM (and now

professor of economics at Harvard) Ken Rogoff
against Bent Larsen in 1976. By abstaining from
f2-f3, White intends to meet ...Nf6-d7 with the
exchange of dark squared bishops, followed by the
rook lift Ra1(f1)-e1–e3-h3, threatening Qd2-h6.
12.Rfd1 e5 13.Be3 Nxe4 14.Nxe4 Bxe4
15.Qxd6 Qh4=;
12.Qe3 e5 13.Bxa7 Nxe4!N 14.Nxe4 Bxe4
15.Rfd1 Bc6! 16.Bc5 dxc5 17.Rxd8 Rfxd8= I’d
rather be Black: ...e5-e4 and ...Bg7-d4 are to
follow.
12...a5
By far Black’s most popular choice.
12...a6!? This was Larsen’s choice against
Rogoff. It appears to be perfectly sound and
deserves further research. 13.Rfe1
(13.b4 b5 14.cxb5 axb5 15.Qe2 Qd7 16.a4
bxa4 17.b5 Bb7 18.Nxa4 Qg4 19.f3 Qg5=)
13...Rb8 14.a4 a5 15.Nd5 Nd7 16.Bxg7
Kxg7 17.Re3 e5 18.Rh3 h5 19.Nc3 Nc5 20.Bc2
Qe7 21.Rd1 Rbd8= 1/2–1/2 (40) Rogoff,K
(2480)-Larsen,B (2625) Biel 1976.
129
15.Rfe1 Once played by a 14-year old
Magnus Carlsen!
(15.Rae1 Nc5 16.Re3 e5! 17.Rh3 h5 18.f4
exf4 19.Qxf4 (19.Rxf4 a4!„) 19...Qe7=)
15...Qb6! Black threatens
...Qb6-d4/...Nd7-c5. If Black manages to exchange
queens he will have the ‘better half’ of an equal
endgame.
a) 15...Re8 16.Re3 Qb6 17.Rh3 Kg8
18.Qh6 Nf6 19.Be2 Kh8 20.g4‚;
b) 15...Nc5 16.Bc2 (16.Bf1 e6 17.Re3 Qf6
18.Rae1 e5 19.Nb5 Rfd8 20.Nc7 Ra7 21.a3 b6
22.Nd5 Bxd5 23.exd5 a4 24.b4 Nb3 25.Qd1 Qg5³
1/2–1/2 (25) Shabalov, A (2520)-Panjwani,R
(2422) Dearborn 2013) 16...b6 17.a3²;
 16.Bf1 This was 14-year old (but still
nearly 2600–rated) Carlsen’s choice.
a) 16.Bc2 Qc5 17.Nd5 (17.Re3 a4 18.Rh3
Nf6 19.Qh6+ Kg8 20.Rd1 axb3 21.axb3 Ra5=)
17...e6 18.Qc3+ e5 19.Rad1 Rfd8 20.Rd3 Nf8!
21.Qd2 Ne6=;
b) 16.Nd5 Bxd5 17.exd5 Qd4 18.Rad1 Ne5
19.Bc2 Qxd2 20.Rxd2 g5! 21.Rd4 Rab8 22.a4
Rh8! 23.f4 gxf4 24.Rxf4 Rbg8 25.Bf5 b6 26.Kf2
Kf8=;
16...a4!
a) 16...Qc5 17.Rad1 Rfd8 18.Re3 e6
19.Qxd6 (19.Rd3 Nf6=) 19...Qxd6 20.Rxd6 Nc5
21.Rxd8 Rxd8 22.f3²;
b) 16...Ne5 17.Nd5 Bxd5 18.exd5 e6
19.Rad1 exd5 20.Qxd5² 1/2–1/2 (65) Carlsen,M
(2581)-Lie, K (2474) Drammen 2004;
17.Qb2 Ne5 18.Nd5 Bxd5 19.exd5 axb3
20.axb3 f6 21.Qc3 g5= In such positions Queens
and Knights tend to cooperate better than Queens
and Bishops.;
13.Rfe1 a4 14.Nd5 Nd7 15.Bxg7 Kxg7
16.Re3 e5 17.Rh3 h5 18.Ne3 Rh8 19.Rg3 Nc5
20.Rd1 h4 21.Rh3 Bxe4µ 0–1 (47)
Kramnik-Grischuk, WCh Candidates KAZ 2011.
13...a4!
130
Here all the games in the database continue
with 14.Nd5 in order to prevent Black’s ...Qd8-a5.
13...Nd7?! 14.Bxg7 Kxg7 15.Re3 Qb6
16.Rh3
(16.Nd5 Bxd5 17.exd5 Rfe8 18.Qc3+ Kg8
19.Rfe1²)
16...h5 17.Rxh5 gxh5 18.Qg5+ Kh8
19.Qxh5+ Kg7 20.Qg5+ Kh8 21.e5‚
14.Nd5
14.f4 Qa5 15.e5 Ng4 16.h3 dxe5 17.fxe5
Nxe5–+
14...Nd7
Now that White has ‘wasted’ a move with
Nc3-d5, Black is in time to counter White’s attack
after the exchange of dark-squared bishops.
14...e6 15.Nb4!
(15.Bb6 Qd7 16.Nb4 (16.Nxf6+ Bxf6
17.Rd1 d5 18.Bb1 d4 19.Qf4 Bg7 20.e5 Bxg2!
21.Rxd4 Qc6 22.Rd6 Qf3 23.Qxf3 Bxf3 24.Re1
Rfc8„ 1/2–1/2 (85) Tomazini,Z
(2411)-Krumpacnik,D (2290) Ptuj SLO 2015)
16...Qe7 17.Bd4 Qc7 18.f4 Nd7 19.Bxg7 Kxg7
20.Kh1 (20.f5 Qb6+ 21.Kh1 Qd4=) 20...Qb6
21.Qc3+ Kg8 22.f5 Ne5 23.f6 Qc5 24.Qd2 Kh8
25.Qh6 Rg8 26.Rf4?? g5–+)
15...Bd7 16.Rd1 Qe7 17.Rfe1 Rfc8 18.Bf1
e5 19.Bb6 Be6
 in many variations prevents Qd2-g5.
17.Rh3 h5 18.Ne3

18.f4 Bxd5 19.exd5 Qa5 20.Qxa5 Rxa5

21.f5 Nf6 22.fxg6 e4 23.Be2 fxg6=;
18.g4?! Bxd5 19.exd5 Nf6 20.gxh5 Nxh5
21.Rxh5 gxh5 22.Kh1 Qf6 23.f4 e4! 24.Qg2+ Kh6
25.Qxe4 Rh8µ
18...Nc5

18...Rh8!? 19.Nd1 Nc5 20.Nc3 Ne6

21.Rd1 Qb6 22.Bf1 Rad8 23.Rd3 Nd4= 1/2–1/2
(32) Espinosa Aranda, A (2399)-Vazquez Igarza,R
(2597) Madrid 2015.
20.b3 19.Rxh5?!
(¹20.Nd5 Nxd5 21.cxd5 Bd7 22.Rc1 f5
23.f3 fxe4 24.fxe4 Qh4 25.Qb4 Bh6 26.Rxc8+
Rxc8 27.Bf2 Qe7 28.Qxb7±)
20...axb3 21.axb3 Nd7 22.Be3 Nc5
23.Qxd6 Qxd6 24.Rxd6 Nxb3= 1/2–1/2 (36)
Wojtaszek,R (2727)-Mamedov,R (2650) Huai’an
2016.
15.Bxg7 Kxg7 16.Re3

16.Qc3+ e5 17.Ne3 Nc5 18.f3 Qa5

19.Qxa5 Rxa5 20.Rd1 Ra6 21.Bb1 Rb6 22.Rf2
Rd8 23.Rdd2 Kf8 24.Rfe2 Ke7 25.Kf1 Ra8
26.Ke1 f6 27.Kd1 Ne6 28.Kc1 Nd4³ 0–1 (64)
Moreno Ibanez,M (2210)-Moreno Ruiz,J (2470)
La Roda 2015.
16...e5! 19...Nxd3! 20.Nf5+ Kf6! 21.Qxd3 gxf5
22.Rxf5+ Ke7µ

0–1 (28) Jerez Perez, A (2405)-Herraiz

Hidalgo, H (2442) Barcelona 2015.

Not only does this move take firm grip

over the d4-square, Black also uncovers his
Queen’s defense of the important g5-square which
131
 5

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7

5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0
Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3

12...Nd7
12...a5 It is hard to say which move order
— 12...a5 or 12...Nd7 is more accurate. In practice
neither is because they will quickly transpose.
13.b3 Nd7 14.Be3 Nc5 transposes to 12...Nd7.
13.Bf2!?
16...Bxd5
(16...Qa5!? This was my choice against
Barbosa. 17.Qe3 (17.b4!? axb3 18.Qb2+ Nf6
19.axb3 Qd8 20.Rxa8 Qxa8=; 17.Qxa5 Rxa5
18.Nxe7 Bxe4µ) 17...Qc5 18.Qc3+ e5?! (18...Nf6!
19.Nxf6 exf6 20.Bf3 b5=) 19.Rad1 Rae8 20.f5
(20.Bg4 Nb6 21.Nxb6 Qxb6 22.Rxd6 Qc5
23.Rfd1 Bxe4=; 20.fxe5! Nxe5 21.b4 axb3
22.axb3±) 20...Bxd5 21.Rxd5 Qc7 22.Qd2 Nf6
23.Rxd6

This is only played about 6% of the time,

13.Be3 being by far the most popular (80%), and
13.Bxg7 second most (13%). However, this move
has at least two clear advantages over 13.Be3:
1) As we will see, it is helpful for the
bishop to be less vulnerable on f2 than it is on e3
in the ...Qb6 lines.
2) In the ...e6/...Be5 lines, a common idea
for Black is to play ...Qd8-h4, which is now
impossible. 23...Rd8! (23...Nxe4 24.f6+ Kh8 25.Qh6
13.b4 Bxd4+ 14.Qxd4 Qb6 15.Qxb6 Rg8 26.Rd3 g5 27.Qxh7+ Kxh7 28.Rh3+ Kg6
Nxb6=; 29.Bh5+ Kh7 30.Bxf7#) 24.c5 Rxd6 (24...Nxe4
13.Bxg7 Kxg7 14.Kh1 This was GM 25.f6+ Kh8 26.Qh6 Rg8 27.Rf3 g5 28.Qxh7+
Oliver Barbosa’s choice against me in 2016. I Kxh7 29.Rh3+ Kg6 30.Bh5+ Kf5 31.Rf3#)
managed to win the game, but both our play could 25.cxd6 Qc6 26.fxg6 hxg6 27.Bf3 Rd8 28.Rd1
have been significantly improved upon. 14...a5 Nxe4 29.Qa5 Qc5!³ Barbosa-Panjwani, NY Int
15.f4 a4 16.Nd5 2016 0–1)
17.exd5 Qa5 18.Qd4+ Kg8 19.Rf3 Qc5
20.Qd3

132
 20...f5! 21.Re3 Nf6 22.Bf3 Rfe8 23.Rae1 (18...Rfd8 19.Rfc1! Preventing Black’s
b5! 24.cxb5 Rab8= freeing ...d6-d5.)
13...a5 14.b3 Nc5 15.Rab1 e6 19.Rfd1 d5! A long forcing variation
results in equality. 20.exd5 exd5 21.Qxd5 Qxe2
15...Qb6 16.a3 Qd8 This has been played a 22.Bxc5 Bf8! 23.Bxf8 Rc2 24.f4 Rd8!! 25.Qf3
few times by Tiviakov and Larsen. Black is sitting Qxf3 26.gxf3 Rxd1+ 27.Rxd1 Kxf8 28.Rd7 Rxa2
back and waiting for White to ‘do something’ like 29.Rxb7 Rb2 30.Ra7 Rxb3 31.Rxa5 Rxb6=
b3-b4 in which case Black at present has adequate 16...Be5 17.Nb5
counterplay. That being said, if White calmly
continues developing then eventually he will
achieve a favorable b3-b4, and by ‘wasting’ time
with ...Qd8-b6-d8 Black’s counterplay has been
stalled. (16...Nxb3?? 17.Qd1!+– Note that if
White’s bishop were on e3, this would not be
possible.) 17.Bd1!?
(17.b4?! axb4 18.axb4 Ra3„)
17...b6 18.Bc2 Qb8 19.Rfc1 Rc8 20.b4²
16.Rfd1

16.Nb5 This is premature; White should

first lure Black’s bishop to e5 (with 16.Rfd1) so
that he can later gain a tempo with f3-f4. Now
Black can do without ...Bg7-e5 because he is in
time to connect his rooks. 16...Bxb5 17.cxb5 Qe7 Compare this with the analogous line after
18.b6 13.Be3. There, 17.Nb5 is just a blunder because of
(18.Rfd1 Rfd8 19.b6 d5=) 17...Qh4!
18...Rfc8! 17...Bxb5 18.cxb5 Qe7!?

18...b6!? This was my choice when I faced

this position over the board. Precise move orders
are not so important here; Black will eventually
play both ...b6 and ...Qe7, then look for
opportunities to break with ...d6-d5. 19.Rbc1 Qe7

133

20.Rc4?!
(20.Qe3 Rad8 21.f4 Bg7 22.g4 Bh6! 23.g5
Bg7 24.Bg4 f6!³)
20...d5! 21.exd5 exd5 22.Bh4 Qc7
23.Qxd5 Bxh2+ 24.Kh1 Rae8 25.Bf1 Re6 26.Qg5
Re5 27.Qf6?? Rh5 28.Rg4
(28.g4 Be5 29.gxh5 Bxf6 30.Bxf6 Qg3
31.Bg2 Ne6 32.Rg4 Qf2–+)
28...Be5 29.Qc6 Qxc6 30.bxc6 f5 31.Rc4
g5–+ 0–1 (37) Findlay,I-Panjwani,R Calgary
2016.
19.Qe3 b6 20.f4 Bg7
20...Bf6!? is also possible.
21.Rd2
We have been following
Macieja-Zaninotto, 2016. Here I recommend an
improvement which turns out to be a fairly
thematic way of handling such positions.
21...Bh6!
134
Since White does not possess a knight to
plant on d5, Black is able to go for the ...e6-e5
push, which also unleashes pressure on the weak
e4-pawn.
21...Rfd8 22.Rbd1
(22.g4!? This would have been a better try
for Macieja. 22...e5 23.f5 Bf6 24.Qf3²)
22...Rac8 23.a3 Bh6! 24.Bg4 Rc7
Zaninotto should have maintained some pressure
on the c-file.
(24...Rb8? 25.b4 axb4 26.axb4 Nxe4
27.Qxe4 f5 28.Qf3 (28.Qe1 fxg4 29.Bh4 g5
30.fxg5 Bxg5 31.Bxg5 Qxg5 32.Qxe6+±)
28...fxg4 29.Qxg4± Macieja-Zaninotto, Najdorf
Memorial 2016)
25.Qf3
(25.b4 axb4 26.axb4 Na4³)
25...e5! 26.b4 axb4 27.axb4 Na4 28.Qb3
Nc3 29.Bxb6 Nxd1 30.Rxd1 exf4 31.Bxc7 Qxc7
32.Qd5 Rb8 33.Be2 f3! 34.gxf3 Bf4=
22.Rbd1 Rad8 23.e5
23.Bf3 e5 24.g3 Rfe8„
23...dxe5
23...d5 24.Rc2 Rc8
(24...f6 25.Bh4 g5 26.exf6 Rxf6 27.Bxg5
Bxg5 28.fxg5 Rf5 29.h4²)
25.Rdc1 f6 26.Bh4 Bg7 27.a3²
24.Qxe5 Bg7 25.Qe3 Rxd2 26.Rxd2
Rd8=
 6 Bxb5 21.cxb5 b6 22.f4 Bg7 23.Bxg7 Kxg7 24.Rb2
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7
5.c4 Nc6 6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0
Bd7 10.Qd2 Nxd4 11.Bxd4 Bc6 12.f3 Nd7
13.Be3 a5
Most people automatically play 14.b3 here,
but it isn’t technically necessary to do so
immediately. If White wants to expand on the
queenside, though, b2-b3 is inevitable, as a direct
a2-a3 will invariably be met by ...a5-a4!
14.b3
14.Rab1 Nc5
(14...a4 15.b4 axb3 16.axb3²)
15.Kh1 GM Fedorowicz played this
against me and then offered a draw the next move.
(15.Rfd1 Be5 16.Bg5 (16.Bd4 Ne6=)
16...Ne6 17.Bh6 Re8 18.b3 Nc5 19.Qe3 e6 20.Nb5
135
24...d5! White’s weak dark squares justify
the following pawn sacrifice to open files and
diagonals. 25.exd5 exd5 26.Qd4+ Kg8 27.Qxd5
Qe7 28.Qf3 Rad8©)
15...a4 16.Rfc1 1/2–1/2
Fedorowicz-Panjwani 2015
(16.Bg5 Ne6 17.Be3 Nc5=; 16.Nd5 e6
17.Nb4 Qe7„)
16...Be5 This is likely how I would have
played on if I had declined Fed’s draw offer.
(16...f5!? It’s nice to always have this
option in a must-win type situation, but in general
I think Black should hold off on ...f7-f5 until he at
least connects his rooks, or until White does
something to disrupt the harmony in his position
(like Nc3-b5).
17.exf5 Rxf5 18.Nd5
(18.b4 axb3 19.axb3 Ra3 20.Bd1 Qf8!
The ‘degenerate’ plan of ...Rf5-h5 and
...Qf8-f6-h4 is actually Black’s best here. 21.Nb5
Ra8 22.b4 Ne4!∞)
18...Qf8∞ This is probably ‘objectively’
+= (whatever that means) but the position is such a
mess that in a practical game both sides have equal
chances.)
17.b4 axb3 18.axb3 Ra3 19.Bd1 e6
20.Bg5
(20.b4? Qh4 21.Bg1 Qf6 22.bxc5 Bxc3
23.Qe2 d5!³)
20...f6
(20...Qd7!? 21.b4 Na4 22.Bxa4 (22.Ne2
b5!„) 22...Bxa4 23.b5 Qc7! 24.f4 Bg7 25.f5 Be5
26.Ne2 The computer calls this equal but I find it a
little awkward to function with my bishop stranded
on a4 so I prefer 20...f6 instead.)
21.Bh4 g5 22.Bg3 Bxg3 23.hxg3 e5
I like the look of Black’s dark-squared
Stonewall formation (shift Black’s g,f,e,d-pawns
to the right one square). 24.b4 Ne6=
14...Nc5
136
The first stage of Black’s plan is complete.
Far and away the most common (nearly 80% of
the time) way to continue for Black is with the
spectacular, if unbelievable ...Qb6, ...Rfc8, ...h5,
...Qd8, ...Kh7, and ...Qh8 with pressure on the long
diagonal. This is the line I mentioned early in the
book, which Mamedov suggested was equal.
Playing this plan on autopilot has its appeal, but
because of its extravagance most White players are
aware of it nowadays, and will not be caught off
guard. The line I recommend in lieu of this is, in
contrast, quite rare (it is played only about 8% of
the time, but it has been growing in popularity)
and in my judgment equally sound, but with far
greater surprise value. The idea is to fight for the
center with ...e7-e6 and ...Bg7-e5, ideally followed
by ...f7-f5. Black’s bishop is rather safe on e5,
because White can scarcely achieve f3-f4 without
losing the e4-pawn. White tends to focus his
attention in one (or more) of three directions:
either he will directly target the weakness on d6
that Black creates after ...e7-e6, or he will try to
achieve f3-f4 which indirectly targets d6 by hitting
its defender (the bishop to be on e5), or as always
White can ‘ignore’ Black’s central gestures and
continue with the usual a3/b4 queenside
expansion.
15.Rab1
This is by far the most popular move,
played nearly 90% of the time. 15.Rfc1 Be5
(15...e6!? Move order is not important
here; ...e6/...Be5 are interchangeable.)
16.Rab1 e6 17.Qe1 Qf6 18.Qd2
(18.g3 Bd4 (18...g5!?) 19.Qd2 Bxe3+
20.Qxe3 e5 21.Nd5 Qd8 22.f4 exf4 23.Qxf4 Bxd5
24.exd5 f5³)
18...Qh4! 19.g3
19...Bxg3
(19...Qe7 GM Maze probably wanted to
avoid a draw (with 19...Bxg3) so chose to play on
this way. 20.Nd1 b6 21.Nf2 Rad8 22.Nd3 Bg7
(22...f5 23.exf5 Rxf5 24.Rf1 Bg7=) 23.Rc2 f5
24.exf5 Rxf5 25.Rf1 Rdf8 26.Ne1 h5 27.Bh6 e5
28.Bxg7 Kxg7 29.Qc3 Ne6µ 0–1 (68)
Makka,I-Maze,S Aix-les-Bains FRA 2011)
20.hxg3 Qxg3+ 21.Kh1
(21.Kf1 f5 22.Bxc5 dxc5 23.Qe3 fxe4
24.Nxe4 Bxe4 25.Qxe4 Rf5! 26.Qxe6+ Kh8
27.Qe3 Raf8–+)
21...f5
22.Bxc5!
(22.Qe1 Qh3+ 23.Kg1 fxe4 24.Bxc5
exf3–+)
22...dxc5 23.Rf1™ Black can give a
perpetual whenever he wants, it’s just a matter of
137
how many accurate moves he wants to make
White find before he gives the draw.
(23.Qe1? Qe5µ)
23...g5 24.Qe1™ 24...Qh3+ 25.Kg1 Rf6
(25...g4 26.Rf2 g3 27.Rg2 f4∞)
26.Rf2 Rh6 27.Rg2 Qh1+ 28.Kf2 Qh4+
29.Kf1 Qh1+ 30.Kf2=;
15.Rac1
The second most popular move. GM
Lenderman played this against me in 2013. My
game against him will be our main line in what
follows.
15...e6
(15...b6 An interesting recommendation of
Khalifman’s which is a perfectly viable alternative
to 15...e6. Black intends the crafty maneuver
...Ra8-a7-d7. I played this against GM Kovalyov
in 2013; I was unsuccessful in that game but it was
not due to the opening. 16.Rfd1
(16.Bd1 Not the last time we will see this
idea. 16...Qb8!?
a) 16...e5?! 17.Bc2 Ra7 (17...Ne6 18.Rfd1
Nd4 19.Bxd4 exd4 20.Ne2 Qh4 21.g3 Qh5
22.Nxd4±) 18.Rfd1 Rd7 19.a3 Ne6 20.b4 f5
21.Nd5 Nd4 22.Bg5±;
b) 16...e6 This deserves further research.
17.a3 a4 18.b4 Nb3 19.Bxb3 axb3 20.Qb2 d5∞
(20...Ba4 21.Rfd1 Qc7 22.Bd4 Bh6 23.Nxa4 Bxc1
24.Qxc1 Rxa4?? 25.Bf6™+–);
17.a3
 17...a4! (17...b5 18.b4 axb4 19.axb4 Ne6
20.Nxb5 Bxb5 21.cxb5 Ra3 22.Re1 Qxb5 23.Be2
Qb7 24.Kh1 Be5 25.Rc2 Rxe3 26.Qxe3 Qxb4²)
18.b4 (18.bxa4 Bxa4 19.Bxa4 Bxc3 20.Qxc3 Nxa4
21.Qd4 Nc5 22.Bh6 Ne6 23.Qc3 Re8 24.f4 f6
25.f5 Nc5 26.fxg6 hxg6 27.Rxf6 exf6 28.Qxf6
Ra7 29.Qxg6+ Kh8=) 18...Nb3 19.Bxb3 axb3
20.b5 Bb7 21.Qb2
21...d5!! 22.exd5 Qd6 23.Qxb3 Rxa3
24.Qb2 Rxc3 25.Rxc3 Rc8=)
16...Ra7
138
17.Bf1 (17.Nb5 Rd7 18.Nd4 Bb7 19.Rb1
e5 20.Nb5 f5„; 17.Nd5 Rd7 18.Bg5 f6 19.Be3 e6
20.Nc3 f5 21.exf5 Rxf5 22.f4 Qh4 23.g3 Qh3
24.Bf1 Qh5 25.Be2=) 17...Rd7 The following is
my game against Kovalyov. 18.Qc2 e5 (18...e6
19.a3 a4 20.b4 Nb3 21.b5 Bb7 22.Nxa4±) 19.a3
Ne6!
This would have led to a very interesting
game. (19...f5?! 20.Nd5 fxe4 21.fxe4 Qh4 22.g3!
Qxe4 23.Qxe4 Nxe4 24.Bg2 Bxd5 25.Rxd5 Nc5
26.Rcd1 Nxb3 27.Rxd6 Rxd6 28.Rxd6± 1–0 (43)
Kovalyov-Panjwani Michigan 2013) 20.b4
(20.Nb5 f5 21.exf5 gxf5 22.Bd3 e4 23.fxe4 f4
24.Bf2 f3©) 20...Nd4 21.Bxd4 (21.Qf2 f5∞)
21...exd4 22.Ne2 axb4 23.axb4 Qg5 24.Kh1 f5©)
16.Rfd1
a) 16.Nb5!? Be5 17.Bg5 Qb6
 18.Qe3 (18.Kh1 Rfe8 19.Nxd6?? Red8–+) (23.Qxd6?? Qg5–+) 23...Bf3!!–+
18...Bxb5! We have come across the resulting Gorgeous.;
structure in the 13.Bf2 line. 19.cxb5 Rac8 20.f4 16...Be5
(20.Rfd1 Rfe8 21.Qf2 Qc7 22.Be3 b6 23.f4 Bg7
24.Bf3 Qe7= Black is ready to continue with
...d6-d5.) 20...Bg7 21.e5 Rfe8 22.exd6 Qxd6
23.Rfd1 Qf8!

White must already take precautions

against Black’s immediate threat of ...Qd8-h4.
17.Bg5 This was Lenderman’s choice
against me. I played correctly for the next 6
Black’s idea is to open up the center with moves, then stumbled — fortunately he didn’t
...e6-e5 and take advantage of White’s weak dark punish my error.
squares and loose pieces. 24.Bc4 (24.b6 e5„; a) 17.Nb5? Qh4 18.g3 (18.h3 Qg3
24.Bg4 h5 25.Bh3 f5! 26.g3 e5!„) 24...b6 25.Bh4 19.Nxd6 b6 20.Nb5 f5–+) 18...Bxg3 19.hxg3
(25.Rc2 Kh8! 26.a3 f5 (26...e5=) 27.Bh4 e5 Qxg3+
28.Bf2 Ne4=) 25...Kh8! Sidestepping the light
diagonal in order to play ...f7-f5 before ...e6-e5.
(25...e5? 26.Rf1!±) 26.Rc2 f5! 27.Rcd2 e5 28.fxe5
(28.Bf2 exf4 29.Qxf4 Be5 30.Qh4 Bf6 31.Qh3
Rb8=) 28...Rxe5 29.Qf4 Re4 30.Qg3 a4 31.Rd8
Re8 32.Rxe8 Qxe8=;
b) 16.Bd4 Be5 17.Rfd1 Qe7 18.f4 Bxf4
19.Qxf4 e5 20.Qg3 exd4 21.Rxd4 Nxe4 22.Nxe4
Bxe4 23.Rxd6?
139

Remember the ensuing ideas, they come up
in various permutations of this type of position.
20.Kf1 (20.Kh1 Nxe4! 21.Qe1 Qh3+ 22.Kg1 Ng3
23.Bb6 Nxe2+ 24.Qxe2 Bxf3–+) 20...Nxe4!
21.Qd3 (21.fxe4 f5!–+) 21...f5!–+ White can’t
untangle, so Black can take his time with ideas
like ...g6-g5-g4.;
b) 17.Bd4 Qe7 18.f4 Bxf4 19.Qxf4 e5
20.Qe3 exd4 21.Qxd4 Nxe4 22.Nxe4 Bxe4
23.Qxd6 Qg5 24.Qg3 Qxg3 25.hxg3 Rfd8=;
c) 17.g3 Qe7 18.Bg5 (18.Nb5 Rfd8 19.Bg5
Bf6 20.Bxf6 Qxf6 21.Nxd6 e5 22.Qe3 Nxb3
23.axb3 Rxd6=) 18...f6 19.Be3 Rad8 (19...b6?
20.f4 Bxc3 21.Qxd6!±) 20.Nb5 f5
21.exf5 exf5!„ I admit it is always hard to
decide what to recapture with on f5. It needs to be
decided on a case-by-case basis. Here, Black’s
a5-pawn is hanging unless Black opens up the
e-file to tie down White’s queen to the defense of
the e3-bishop. (21...gxf5 22.Bg5 Bf6 23.Bh6 Rf7
24.Qxa5±);
140
d) 17.Rc2 Witness how even a top
grandmaster can fall victim to Black’s sudden
counterplay. 17...Qe7 18.Nb5?
18...Qh4! 19.g3 Bxg3 20.hxg3 Qxg3+
21.Kf1 Nxe4 22.Qd3 f5 23.Nd4 f4 24.Nxc6 fxe3
25.Qxe3 Rf4 26.Nd4 Rh4 27.Qg1 Rh1 0–1
Lupulescu-Iturrizaga, Baku Olympiad 2016;
17...Qb6 18.Qe3
(18.Kh1 a4 19.Be7 axb3 20.Bxf8 Kxf8
21.axb3 Nxb3 22.Qb2 Qb4 23.Rc2 Bg7 24.Rb1
Ra3–+)
18...a4! 19.Nb5
(19.b4 Qxb4 20.Be7 a3 21.Bxf8 Rxf8©;
19.f4 Bxc3! 20.Rxc3 axb3 21.axb3 Qa7 22.e5 Ne4
23.Qxa7 Rxa7 24.Re3 Nxg5 25.fxg5 d5=)
19...axb3 20.axb3 Ra2 21.Bf1 Rb2! 22.Rb1
Rxb1 23.Rxb1
23...Ra8?!
(¹23...Bxb5 24.cxb5 Ra8 25.f4 Ra1
26.Rxa1 Bxa1 27.b4 Na4³)
24.f4 Ra1 25.Rxa1 Bxa1 26.Nxd6?
 (26.b4! Bxb5 (26...Na4 27.Qxb6 Nxb6
28.Nxd6+–) 27.bxc5 dxc5 28.Qb3!±)
26...Qxb3 27.Qxb3 Bd4+ 28.Kh1 Nxb3³
1/2–1/2 Lenderman-Panjwani, Arlington 2013;
15.Rfd1
White can only temporarily delay ...e7-e6
with this move, as after 15...Be5 White has no
appealing way of preventing it. 15...Be5 16.Bg5?!
(16.Bd4 Ne6=; 16.Rab1 transposes to the
main line 15.Rab1.)
16...Qb6 17.Kh1 f6 18.Be3
(18.Bh6 Rfb8! 19.Be3 a4 20.Rab1 axb3
21.axb3 Qb4 22.Nd5 Qxd2 23.Rxd2 Bxd5 24.cxd5
Ra3³)
18...a4 19.Rab1 axb3 20.axb3 Qb4!
Typical for this line. Black sets up a
dark-square blockade, to be followed by ...Ra8-a3
if unhindered. 21.Nd5 Qxd2 22.Bxd2
(22.Rxd2 Bxd5 23.cxd5 Ra3 24.Bc4 Rb8=)
22...Bxd5 23.cxd5 Ra3 24.Bc4 b5 25.Bxb5
Rxb3 26.Rxb3 Nxb3 27.Be3 Rb8 28.Be2 Nc5=
141
15...e6
16.Rfd1
16.Rfc1 Be5 17.Bf1
(17.g3 h5 18.Nb5 h4 19.g4 f5„)
17...f5
(17...Qe7!? 18.Rc2 f5 19.exf5 gxf5 20.Re1
b6 21.Ne2 Qh4 22.Bf4 Rad8 23.Rcc1 Rd7 24.Bg3
Qf6 25.Rcd1 f4 26.Bf2 Qg6 27.Kh1 Rg7 28.Nc3
Qh5„ 0–1 Perera, P-Cuenca Jimenez,J Seville ESP
2004)
18.exf5
It is a perpetual dilemma for this line —
what to recapture on f5 with? I have yet to come
up with a general algorithm; the concrete features
of the position usually suggest one over the others,
but here there are two viable continuations.
18...Rxf5 Black has two bishops and a rook
pointed at White’s king, and his queen is ready to
join the action with ...Qd8-h4.
 a) 18...gxf5!? 19.Ne2 (19.f4 Bxc3 20.Qxc3
Ne4 21.Qd4 Rf6 22.a3 Rg6„) 19...Qh4 (19...Rf7
20.Nd4 Rg7 21.Nxc6 bxc6 22.g3± 1/2–1/2 (49)
Bauer,C-Schmitt,A Clichy FRA 2004) 20.Nf4
Qh6„;
b) 18...exf5? 19.Nd5±;
19.f4
a) 19.a3?? Qh4 20.g3 Bxg3 21.hxg3
Qxg3+ 22.Bg2 (22.Qg2 Qe5–+) 22...Rxf3 23.Bxc5
dxc5–+ ...Ra8-f8 to follow.;
b) 19.Nb5 Qh4 20.g3 Bxg3 21.hxg3
Qxg3+ 22.Bg2 Bxf3 (22...Rxf3 23.Bxc5 dxc5
24.Rf1 Rf5 25.Rxf5 exf5 26.Qf2 Qg4©) 23.Nxd6
Bxg2 24.Nxf5 exf5 25.Bf4 Qg4 26.Qxg2 Qxf4
27.Qd5+ Kg7 28.Qxc5 Qg3+ 29.Kf1 Qh3+ 30.Kf2 22...Rxf4! 23.gxf4
Qh2+ 31.Kf1 Qh3+ 32.Kg1=; (23.Qxf4? Qb6–+)
19...Bg7 23...Qb6©;
16.Bd1

A point made by GM Rowson in his book

Chess for Zebras: a pawn on f4 weakens the king
in a way which should not be underestimated. It is
such a common move in so many openings that we
take it for granted, but it weakens squares around
White’s king and here Black is ready to shine light
on those weaknesses with ...e6-e5. 20.g3
(20.Be2 e5! 21.Rf1 exf4 22.Rxf4 Rxf4
23.Bxf4 Qf6 24.Nd5 Qd4+ 25.Be3 Qxd2 26.Bxd2
Bxd5 27.cxd5 Ne4 28.Be1 Bd4+ 29.Kh1 Rc8³)
20...e5! 21.Bh3
(21.fxe5 Qe8!„)
21...exf4 22.Bxf4
142
I first came across this move when Cuban
GM Abreu seemed to come up with it over the
board against me after a long think in the game I
mentioned in the introduction of this book.
However, he was not the first to play it; that honor
goes to the late Bulgarian GM Georgi Tringov
who played it in 1986.16.Bd1 is a perfectly logical
move: the bishop will relocate to c2, from where it
simultaneously bolsters the critical e4, b3, and a4
squares. White already threatens 17.a3 since the
knight cannot hop into b3 after 17...a4 18.b4. In
the last few years, 16.Bd1 has been played more
times than it has in all previous years combined. I
think the most common move, 16...Be5, does not
equalize for Black, but 16...Qc7 does. It makes
sense too — Black targets the c4-square which has
been abandoned by the relocation of the e2-bishop.
I thank IM Dave Ross for suggesting this move to
me. 16...Qc7! This move is a good illustration of
the aforementioned ‘reactive’ nature of Black’s
play in this line. White relocates his bishop to c2,
so Black targets the c4-pawn with his queen, and
in some lines with his knight by relocating to e5
via d7.
a) 16...f5?! 17.exf5 gxf5 (17...Rxf5
18.Ne2± 1–0 (43) Tringov, G-Haik,A Vrnjacka
Banja 1986) 18.Bc2 f4 19.Bf2 Qg5 20.Kh1 Be5
21.Rfe1 Qg7 22.Ne2 b6 23.Nd4 Bb7 24.a3 Rf6
25.b4 Nd7 26.Nb5 Kh8 27.Bg1 axb4 28.axb4 Rg8
29.Re2 d5 30.Rbe1 (¹30.cxd5 Bxd5 31.Bb3 Bxb3
32.Rxb3±) 30...dxc4 31.Nd4 Rh6 32.Rxe5?? Nxe5
33.Rxe5 Qxe5 34.Nf5 Rf6?? (34...Bxf3!–+) An important branching point. White’s
35.Nd6 (¹35.Bd4 Rd8 36.Bxe5 Rxd2 37.Bxf6+ a2-a3 is imminent; other plans do not make sense
Kg8 38.Nd4=) 35...Qd5 36.Qc3 e5 37.Nxb7 Qxb7 of White’s bishop maneuver. It is just a question
38.Qxe5 Qg7 39.g4 fxg3 40.Bd4 g2+ 41.Kg1 Rxf3 of how White chooses to position himself before
0–1 Abreu Delgado-Panjwani, US Masters 2013; playing it. 18.Rfd1
b) 16...Be5 17.a3! Qc7 (17...f5 18.exf5 Qf6 a) 18.Rfc1 Nd7! White’s last move (by
19.Ne2 exf5 20.g3!²) 18.b4 axb4 19.axb4 Na4 X-ray) defended the c4-pawn, so Black calls the
20.Nxa4 Bxa4 21.Be2 Rfd8 22.f4 Bg7 23.b5 d5 knight for reinforcement. 19.a3 Ne5 20.Nb5 Qe7
24.cxd5 Qc2 25.Rfc1 Qxd2 26.Bxd2 Bd4+ 27.Kf1 21.Bb6 Rd7 22.f4 (22.Qe2 Bh6 23.Rd1 d5 24.Bd4
exd5 28.e5²; dxc4 25.bxc4 Bg7³) 22...Bxb5 23.cxb5 Ng4
17.Bc2 24.Bd1 Nf6 25.Bf3
(17.a3 Rfd8 18.b4 axb4 19.axb4 Na4
20.Nxa4 (20.Bxa4 Bxa4 21.b5 Qxc4 22.Rfc1 Rdc8
23.Nd5 exd5 24.Rxc4 dxc4„) 20...Bxa4 21.Be2

25...Bh6! 26.g3 e5„;

b) 18.a3 a4! 19.Nxa4 (19.b4 Nb3 20.Bxb3
axb3 21.Rxb3 Be8!=) 19...Bxa4 20.bxa4 Rdc8=;
21...d5! 22.exd5 exd5 23.cxd5 Qe5=) c) 18.Bg5 Rdc8 19.Rfd1 Na6! 20.a3
17...Rfd8

143

20...b5!! 21.cxb5 Be8 22.bxa6 Rxa6
23.Bd3 Bxc3 24.Qf2 Raa8=;
18...Be5 19.a3
(19.g3 f5 20.exf5 gxf5 21.Nb5 Qg7„)
19...a4!
A major idea of 17...Qc7 is to be able to
meet a2-a3 with this move, softening the c4-pawn
further. 20.b4
a) 20.bxa4 Rdc8 21.Rb4 Na6 22.Nb5 Bxb5
23.Rxb5 Qxc4 24.Bb3 Qc6=;
b) 20.Nxa4 Bxa4 21.bxa4 Rdc8 22.f4 Bf6
23.Qb4 (23.Qxd6 Nxa4 24.Qxc7 Rxc7=) 23...Qc6
24.Bxc5 Qxc5+ 25.Qxc5 Rxc5 26.Bb3 Rc6
27.Rd3 Bd8=;
20...Nb3! It will be evident from this
variation why the queen is so valuable on c7 in
this line. 21.Bxb3
(21.Qd3 d5„)
21...axb3 22.Rxb3 Be8! 23.Qe2 Rdc8 24.f4
Bf6 25.Nb5 Qxc4 26.Qxc4 Rxc4 27.Nxd6 Ba4!
28.Nxc4 Bxb3 29.Rc1 Bxc4 30.Rxc4 Rxa3
144
31.Rc8+ Kg7 32.Bc5 Bc3=;
16.Nb5
We saw the following idea in my game
against Lenderman after 15.Rac1. 16...Be5 17.Bg5
Qb6 18.Qe3 White threatens f3-f4 and d6 will fall,
so Black must act quick.
(18.Kh1 Rfe8 19.Rbd1 (19.Nxd6?? Rad8
20.Bxd8 Rxd8–+) 19...d5! 20.Qe3 (20.Be3 Red8=)
20...Bxb5 21.cxb5 Bg7 22.e5 Nd7 23.Qxb6
Nxb6=)
18...Bxb5! 19.cxb5 Rac8
20.Kh1
a) 20.f4 Bg7 21.Rfd1 Rfe8 22.Rbc1 d5
23.e5 Bf8 24.Rc3 (24.Bh4 Nd7=) 24...Nd7
25.Qxb6 Nxb6 26.Rdc1 Be7 27.Bh6 Bf8 28.Rxc8
Rxc8 29.Rxc8 Nxc8 30.Bxf8 Kxf8=;
b) 20.Rfd1 Rfe8 21.Qf2 Qa7 22.Bc4 (22.a3
b6 23.Be3 Qb7 24.Bd4 Bxd4 25.Qxd4 d5!³)
22...b6 23.g3 (23.Rbc1 Qb8= White has no
constructive plan, whereas Black has long term
ideas of ...f7-f5 or even ...d6-d5 after
...Rc8-c7-d7/...Qb8-b7.) 23...f5 24.exf5 d5 25.Be2 28...gxf5 29.bxa4 d5 30.Ba2 Bxh2 31.a5
gxf5 26.Qe3 Qg7„; Bd6 32.axb6 Rg7„;
20...Rfe8 16.Bg5?? Bxc3–+
16...Be5

21.Qd2 White targets Black’s weak

d6-pawn, but does not demonstrate any
constructive plan. Black has time to regroup and
prepare for ...d6-d5 or ...f7-f5. The following is
just one illustration of how Black can successfully
regroup.
a) 21.f4 Bg7 22.Rbc1 (22.a3 d5 23.e5
Na4!³) 22...d5 23.e5 Nd7 24.Qxb6 Nxb6 25.Bh4
Bh6³;
b) 21.Rfd1 d5„;
21...Qa7! 22.Rfd1 b6 23.Rbc1 Qb8 24.Bc4
Rc7 25.Qe2 Rd7 26.Rd2
The database reveals that it was Danish IM
Jan Sorensen who introduced the idea of
...e6/...Be5 into tournament practice. In 1990 and
1991 he scored 1.5/2 against then IMs (now GMs)
J. Kristiansen and A. Fishbein. It has since been
championed most notably by Azerbaijani GM
Gadir Guseinov, and has also been implemented
by Grandmasters Iturrizaga, Bauer, Leon Hoyos,
Cebalo, Lie, Perelshteyn, and many others.
17.Qe1

17.Bf2 The fact that this retreat is played

so often indicates to me that the bishop should go
directly to f2 from d4 earlier on. 17...Qe7 18.Rbc1
f5

26...f5! 27.a3
(27.exf5 gxf5 28.Be3 d5 29.Bxc5 bxc5
30.b6 Rde7 31.Bb5 Rd8 32.Rxc5 Bxh2 33.Rdc2
Bg3=)
27...a4 28.exf5
(28.b4 Nb3 29.Bxb3 axb3µ) 19.exf5 Rxf5!
145
 (19...gxf5 20.Nb5 (20.Bd4 Bxd4+ 21.Qxd4
Rad8 22.Rd2 e5 23.Qf2 Ne6 24.Re1 Kh8 25.Bf1
Qg7„ 0–1 (39) Jakovljevic,V-Leon Hoyos,M
Cento 2011) 20...Rfd8 21.Nd4²)
20.Nb5
(20.Bd4 Bxd4+ 21.Qxd4 Rd8= Black has
ideas of ...e6-e5 and ...Nc5-e6-d4. Black’s
weakened d5-square is less of an issue than
White’s weakened d4-square because Black has a
light-squared bishop to guard his weak square.)
20...Bxb5 21.cxb5 Raf8! 22.Qxa5
22...Ne4! 23.Qe1
(23.Be3?? Bxh2+ 24.Kxh2 Qh4+ 25.Kg1
Rh5–+)
23...Qg5 Black has enormous pressure. The
following is just an illustration of how things
could continue. 24.Kh1 d5 25.g3 Rxf3 26.Bxf3
Rxf3 27.Rc8+ Kg7 28.Bd4 Bxd4 29.Rxd4 Qf6
30.Rc7+ Kh6 31.Rd1 Rf2 32.Qe3+ g5 33.Kg1 Qf5
34.Rxb7 Qh3 35.Qxf2 Nxf2 36.Kxf2 Qxh2+
37.Kf1 d4! 38.Rd7 e5–+;
17.Bd4
146
Black has (at least) two adequate responses
to this move. I prefer 17...Qe7 but the alternative is
more direct and simplifying. 17...Qe7
(17...Bxd4+ 18.Qxd4 Qb6!? Black twists
and turns in order to achieve ...e6-e5 and
...Nc5-e6. 19.a3 e5 20.Qe3 Qd8 21.b4 axb4
22.axb4 Ne6 23.b5 Be8 24.Nd5 Kg7 25.f4
(25.Qb6 Qxb6+ 26.Nxb6 Rd8 27.Ra1 f6 28.Ra7
Nc5 29.Na4 Nxa4 30.Rxa4 Rf7 31.Ra2 (31.c5
Rfd7=) 31...Kf8 32.Rad2 Rfd7=) 25...f6 26.fxe5
fxe5 27.Bg4 Nc5= 0–1 (41) Porat,M-Bejtovic,J
Prague 2011).
18.Bf1
(18.f4 Bxf4 19.Qxf4 e5 20.Qg3 exd4
21.Rxd4 Nxe4 22.Nxe4 Bxe4 23.Re1 f5 24.Bf3
Qe5=)
18...Rad8 19.Qe3 Bxd4 20.Rxd4 e5 21.Rd2
f5
(21...Ne6=)
22.exf5 gxf5 23.f4
23...Ne6! 24.fxe5 f4 25.Qe1 dxe5 26.Rxd8
Qc5+ 27.Qf2 Rxd8³ Bok, B-Lie,K Wijk aan Zee
NED 2010;
17.g3 A recommendation of Khalifman’s
in his (eternally informative) repertoire series for
White. 17...Qe7 18.Nb5 Rfd8
19.Nd4
(19.Bg5 f6 20.Be3 g5 21.Nd4 Be8„ Black
has ideas of ...h7-h5, ...g5-g4, ...Be8-g6. White is
solid but already on the defensive.)
19...d5! 20.cxd5 exd5 21.Qc1 Ne6 22.Nxc6
bxc6 23.Qxc6 d4 24.Bf2 Rac8
25.Qa6
(25.Qa4 Ng5! 26.f4?? Nxe4 27.fxe5
Nc3–+)
25...Rc2 26.Bf1 Qg5!
(26...Rxa2 27.Ra1 Rxa1 28.Rxa1 d3
29.Rd1 Qb4 30.Rxd3 Rxd3 31.Qxd3 Bd4=
1/2–1/2 (46) Nyback-Sorensen 2008)
27.a4?
(27.Qxa5?? Rxf2–+; ¹27.Bd3 Rxa2 28.Ra1
Rb2 29.Rxa5 Nf4„ Computers give a way for
147
White to escape with a draw but it would be
terrifying to defend this over the board.)
27...h5 28.h4 Qf6 29.Rd3 g5–+;
17.Bf1? The Accelerated Dragon is full of
landmines. Grandmasters Chuchelov, Erenburg,
Zubarev, and Beim have misstepped onto the
following. 17...Qh4!
18.h3
(18.g3 Qf6µ The f3-pawn falls.)
18...Qg3 19.Bd3 f5 20.exf5 Nxd3 21.Qxd3
Bxf3–+ Shvayger,Y-Socko,M, Baku 2016;
17.Nb5?
Victims of the following include IM Danny
Rensch and GM Peter Heine Nielsen! 17...Qh4!
18.g3 Bxg3 19.hxg3 Qxg3+ 20.Kf1 Nxe4! 21.fxe4
f5µ;
17.a3
 Black’s last move accompanied a threat of
...Qd8-h4, which is enough for equality, but it is
far more effective if White’s knight has been
driven away from c3. 17...Qf6!?
(17...Qh4!? 18.g3 Bxg3 19.hxg3 Qxg3+
20.Kf1 Qh3+ 21.Ke1 Qh4+ 22.Bf2 Qh1+ 23.Bf1
Qxf3 24.Qe3 Qxe3+ 25.Bxe3 Nxe4 26.Nxe4 Bxe4
27.Bd3 Bc6=)
18.Bd4
(18.Nb5? Qh4! 19.g3 Bxg3 20.hxg3
Qxg3+ 21.Kf1 Nxe4! 22.fxe4 f5–+)
18...Bxd4+
(18...a4 19.Bxe5 dxe5 20.b4 Nb3 21.Qe3
Nd4 22.b5 Bd7∞)
19.Qxd4 Qxd4+ 20.Rxd4 Rfd8 21.b4
In the database Black plays his Queen to f6
about equally as often, but I tend to prefer my
Queen on e7 as a general rule unless there is
something concretely favorable about placing it on
f6.
17...Qf6 18.Nb5
(18.Rbc1 Rad8 19.Bf1 g5„)
18...Rfd8 19.Nd4 Bf4 20.Bf2²
18.a3
18.Nb5 f5 19.exf5 exf5! 20.Qd2?

(20.Bd4 Ne6=)
20...Ne4!! 21.Qe1
(21.fxe4 Qh4–+)
21...Bxb5 22.fxe4 Bc6 23.exf5 Bxh2+
(21.Bd1 e5 22.Rd2 Ne6=) 24.Kxh2 Qxe3µ
21...e5! 22.Rd2 axb4 23.axb4 Ne6 24.b5 18...f5 19.exf5
Be8 25.Nd5 Nd4=
17...Qe7

148
 It is usually a challenge to decide how to 23.fxe4
recapture on f5 but here Black has an
embarrassment of riches — not only are all three 23.Bd3 Rxf3! 24.gxf3 Qg5+ 25.Qg3 Qe3+
recaptures playable, Black can even opt for none 26.Qf2 Qg5+=
of the above with 19...Qf6!?. 23...Nxe4 24.Bd3 Re5 25.Bxe4 Bxe4
19...Rxf5 26.Nxe4 Rxe4=

19...exf5 20.Qd2 Rfe8 21.Nd5 Qf7 22.Bf2

Bxd5 23.Qxd5 Bxh2+ 24.Kxh2 Rxe2 25.Bg3
Qxd5 26.Rxd5 Rd8 27.Bxd6 Ne6©;
19...gxf5 I would recommend this the least
of the four options. 20.b4
(20.Bd4 Qg7 21.Qf2 Bxd4 22.Qxd4 Rfd8
23.Qxg7+ Kxg7 24.b4 Na4 25.Nxa4 Bxa4=)
20...axb4 21.axb4 Qg7 22.Nb5 Bxb5
(22...f4 23.Bf2 Na4 24.Nd4 Bd7 25.Bf1
Kh8 26.Kh1 Rg8 27.Rbc1²)
23.bxc5 Ba4 24.Rxd6! Bxd6 25.cxd6∞;
19...Qf6!? 20.Nb5 Qxf5 21.Nxd6
(¹21.h3 Bxb5 22.cxb5 Rac8∞)
21...Bxh2+ 22.Kxh2 Qe5+ 23.Kg1 Qxe3+
24.Qf2 Qg5=
20.Bd4

20.Nb5 Bxb5 21.cxb5 d5„

20...Bxd4+ 21.Rxd4 e5 22.Rdd1

Black can continue ‘normally’ with

22...Ne6, especially if he wants to keep tension
and play for a win, but the simplest path to
equality is to liquidate things with 22...e4! 22.Nd5
Qg7 23.Rdd1 e4 24.fxe4 Nxe4 25.Bf3 Bxd5
26.Bxe4 Bxe4 27.Qxe4 Re5 28.Qd3 Qe7=
22...e4!

149
 CHAPTER 6

4.Qxd4 VARIATION

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Qxd4 Nf6

The Hyper Accelerated Dragon move order

(2...g6) avoids the Bb5 Sicilian at the cost of
allowing 4.Qxd4 here. In my opinion, the tradeoff
is very much worth it: the Bb5 Sicilian is a real
buzzkill for the dynamic and ambitious Black
player, whereas the 4.Qxd4 line is for the most
part rich in dynamical possibilities and offers
Black real chances to play for a win.
4...Nf6 5.Nc3

We cover White’s main move, 5.e5, in the

next subchapter. This line has some venom, and in
Contents my judgment Black does not equalize in the most
popular way of dealing with it.
5.Bb5 I have faced this move several times
1. 5.sidelines in online blitz but never over the board. It’s
2. 5.e5 basically just a cheapo attempt: if Black plays
5...Bg7 then 6.e5! is strong for White. However,
this unrespectable move actually has, according to
1 Donaldson and Silman, a highly respectable
originator — David Bronstein. 5...a6 (5...Bg7?
6.e5±) 6.e5 This can be a little scary the first time
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Qxd4 you see it, but it’s just a bluff by White. It works
well against an unprepared opponent in 1–minute
chess though!
a) 6.Ba4 b5 7.Bb3 Nc6 8.Qd3 Bg7 9.Nc3
0-0 10.0-0 d6

150
 We have here just a normal Dragon
structure for Black where White’s pieces are
misplaced. 11.Bg5 h6 12.Be3 (12.Bxf6 Bxf6 13.e5
Bf5µ; 12.Bh4 g5 13.Bg3 Nh5³; 12.Bd2 Nd7„)
12...Ng4 13.Bf4 Bb7=;
b) 6.Qa4?! Nxe4! (6...b6 7.e5 Bb7∞
Arribas Lopez-Leon Hoyos, New Orleans 2016)
7.Qxe4 (7.Bxd7+ Bxd7 8.Qxe4 Nc6³) 7...axb5
8.Qe5 f6 9.Qxb5 Qa5+ 10.Qxa5 Rxa5³;
6...axb5 7.exf6 Nc6 8.fxe7
(8.Qd3 b4! 9.0-0 e6³; 8.Qh4 exf6 9.0-0 d5
10.Re1+ Be6 11.Nc3 b4 12.Nb5 Bg7 13.Bh6 0-0
14.Bxg7 Kxg7 15.Nbd4 Qd6µ)
8...Qxe7+ 9.Qe3
Donaldson and Silman give preference to
White here but it is Black who is better. 9...Qxe3+
10.Bxe3 Bg7 11.c3 b4!³
5...Nc6 6.Qa4 d6 7.e5 dxe5 8.Nxe5 Bd7
151
9.Nxd7
Here 9...Qxd7 is the most popular move by
a 6:1 margin but it looks to me that Black
struggles to equalize there. Giving up the two
bishops is already a concession Black has to live
with in this line; he should rush to finish his
development with 9...Nxd7 rather than spend
several tempos on queen moves as he must after
9...Qxd7.
9.Nxc6 Donaldson and Silman give this as
+= however... 9...Qb6!N A nice little novelty
which renders 9.Nxc6 harmless. We wait to see
where White puts his f1–bishop before we decide
how to take back on c6.
(9...Bxc6 10.Bb5² Black is slow getting
developed and castled.)
10.Qb3
(10.Bb5 bxc6! 11.Bc4 Bg7 12.0-0 0-0
13.Bb3 c5 14.Qc4 Rac8„)
10...Bxc6 11.Qxb6 axb6 12.f3 Bg7 13.Be2
0-0 14.Be3 Nd7
 The Nd7/Bc6/b7 are a common trio in the This new move (with the idea of castling
Accelerated Dragon — all three pieces are queenside) was tried by GM Popilski in 2016. It
mutually defended and they exert control over has some bite, so Black should be accurate here.
important central squares. 15.Kf2 Bxc3! 16.bxc3 10...Bg7
Ra3„ (10...Nb6!? 11.Qe4 Bg7 12.0-0-0 0-0 13.h4
9...Nxd7! Qd4=)
11.0-0-0 0-0
9...Qxd7 10.Be3 Bg7 11.Rd1 Qg4 a) 11...Qc8 12.h4 h5 13.Rg1!? (13.Qf4!?‚)
(11...Qc8 12.Be2 0-0 13.0-0 a6 14.Qf4 Qf5 13...Nb6 14.Qe4 0-0 15.g4 hxg4 16.h5‚
(¹14...e6 15.Bf3 Rd8 16.g3²) 15.Qxf5 gxf5 16.Na4 b) 11...Nb6!? tr. 10...Nb6;
e6 17.c4 Rfd8 18.Bb6 Rxd1 19.Rxd1± 1–0 (67) 12.h4!
Rausis,I (2600)-Turner,M (2493) Fagernes NOR (12.Bb5 Nb6 13.Qe4 Qc7=
2015) Popilski-Perelshteyn, Dallas 2016)
12.Qb3! 12...Nb6!
(12.Rd4 Qf5! (12...Qc8 13.Bb5 a6 (12...h5? 13.g4±)
14.Bxc6+ Qxc6 15.Qxc6+ bxc6 16.Ke2 0-0 13.Qe4 Qd4 14.Be3
17.Rhd1 e6 18.Rc4 Nd5 19.Bd4 Rfc8 20.Bxg7
Kxg7 21.Rc5 Rab8 22.b3² 1–0 (43) Erdos,V
(2651)-Gledura,B (2395) Zalakaros HUN 2014)
13.Bd3 Qe5 14.Rb4 Qc7 15.Bb5 0-0 16.Rc4 Rfc8
17.0-0 e6 18.Bf4 Qb6 19.Be3 Qc7= 0–1 (36) Wei
Yi (2706)-Bu Xiangzhi (2681) China CHN 2015)
12...0-0 13.h3 Qb4 14.Be2 Qxb3 15.cxb3
Rfd8 16.Bf3 a6 17.Ke2 Rxd1 18.Rxd1 Rd8
19.Rxd8+ Nxd8 20.b4²
10.Bb5

10.Bd2

14...Qb4!
(14...Qxe4 15.Nxe4 Rfd8 16.Rxd8+ Rxd8
17.Bxb6 axb6 18.c3²)
15.Qxb4 Nxb4 16.h5
(16.a3 Bxc3 17.bxc3 N4d5=)
152
 16...Rfd8=
10...Bg7!

13...Nb6! 14.Nxb6 axb6 15.Qc4 Rd8

A temporary pawn sacrifice.
11.0-0
11.Bxc6 bxc6 12.0-0
(12.Qxc6 Rc8 13.Qf3 Bxc3+! 14.bxc3 Qa5
15.0-0 0-0 16.Re1 Rfe8 17.Bd2 Ne5= White will
not be able to hold on to his extra pawn.)
12...0-0 13.Qxc6 Rc8 14.Qa4
16.Rxd8+ Qxd8 17.c3 Na7 18.Be3 Rc8 19.Qb3
Nxb5 20.Qxb5 Rc6= Black’s pawns are doubled
but White cannot make use of his queenside
majority without allowing Black to undouble
them.
12...Nb6
12...bxc6!? This will transpose to 11.Bxc6.
13.Qh4 bxc6 14.Bh6 e6=
1/2–1/2 (44) Gorovets,A
(2551)-Panjwani,R (2393) New York 2016.

14...Bxc3! 15.bxc3 Qc7 16.Be3 Nb6

(16...Nc5 17.Qh4 Ne6=)
17.Qa5 Qxc3 18.Bxb6 axb6 19.Qxb6
Qxc2=
11...0-0 12.Bxc6

12.Rd1 Qc8 13.Nd5

153
 2 11.Nxe5 Nxe5µ
6...Nd5 7.Qe4

1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Qxd4 Nf6 7.Qb3 Nc7!

5.e5 (7...Nb6 8.a4 a5?? 9.Be3+–)
8.Nc3
If White is going to seek an advantage in (8.Bc4 Ne6=)
this line, this is his best hope. 8...d5 9.exd6 Qxd6 10.Bc4 Be6 11.Bxe6
5...Nc6 Qxe6+ 12.Be3 Qxb3 13.axb3 Bg7=
7...Nb6!

6.Qa4
Black’s knight is less vulnerable on b6 than
6.Qc3 Ne4! it is on b4 in the 7...Ndb4 line, and it is also better
(6...Nd5!?) placed on b6 than c7 because it deters White’s
7.Qe3 d5 8.exd6 Nxd6³; Bf1–c4. What I like most about this line for Black
6.Qh4? Nxe5! 7.Nxe5 Qa5+ 8.Nc3 is that White has to be extremely accurate to not
Qxe5+³; immediately be worse — there are a lot of ways to
6.Qe3? Ng4 7.Qe4 d5! go astray.
7...Ndb4?! This is the most popular move,
even though Sutovsky basically refuted the system
ten years ago. 8.Bb5 Qa5 9.Nc3

(7...Ngxe5 8.Nxe5 Qa5+ 9.Nc3 Qxe5

10.Qxe5 Nxe5 11.Nb5 Kd8 12.Be3©)
8.exd6 Bf5 9.Qe2 Qxd6 10.h3 Nge5
154
 9...Bg7 8...Bg7
(9...d5 10.exd6 Bf5 11.Qe5 Nxc2+ 12.Ke2
0-0-0 13.Qxh8 Nxa1

9.Bf4
This used to be thought of as better for
Black (for example Donaldson and Silman say 9.Bg5?! d5! 10.exd6 Bxc3+ 11.bxc3 Qxd6
“Black has a distinct advantage”) but the optics are 12.Bb5 f6 13.Bf4 e5³
misleading: White has a clear advantage. 14.Qxh7 9...d5!
Nc2 15.Rd1±)
10.0-0 0-0 11.a3 d5 12.exd6 Bf5 As we have seen, this is the central motive
for Black in this variation. 9...0-0 10.0-0-0 d5
11.exd6 Bf5? (¹11...Bxc3 12.bxc3 Bf5∞) 12.dxe7±
10.exd6 Bf5

13.axb4! Qxa1 14.Qh4 exd6 15.Nd5 f6

16.Bxc6 bxc6 17.Ne7+ Kh8 18.Nxf5 gxf5
19.Nd4± 1–0 Sutovsky,E (2637)-Guseinov,G
(2586) Dresden GER 2007.
8.Nc3 White might be well advised to already
seek to liquidate the position into an equal
8.Bb5?! a6 9.Bxc6 dxc6 10.0-0 Bg7 endgame after 11.dxe7. Otherwise, with Black’s
11.Qh4 h6 12.Nc3 Bf5³; pieces so active and White’s king still in the
8.Bf4?! d5 9.exd6 Bf5 10.Qe2 Nd5! center, things could become dire pretty fast on the
11.Bg3 slightest misstep by White.
(11.Be5 f6 12.Bg3 e5µ) 11.Qe3
11...Ndb4 12.Na3 Bg7µ
155
 11.dxe7 (This used to be the most popular
move but in the last few years it has been eclipsed
by 11.Qe3 and 11.Qe2.) 11...Bxc3+!
(11...Qd7 12.Qe2 (12.Qe3 Nb4 13.Rc1
Nxc2+ 14.Rxc2 Bxc2 15.Bb5 Bxc3+ 16.Qxc3
Qxb5 17.Qxh8+ Kxe7=) 12...Nb4 13.Rc1 Qxe7
14.Bd6! Qxe2+ 15.Bxe2 Nxc2+ 16.Kf1 Nd4
17.Nxd4 Bxd4 18.g4 Bd7 19.Kg2±)
12.bxc3 Bxe4 13.exd8=Q+ Rxd8
14.Bh6!
(14.Bg5 Rd5 15.Bh6 Rh5 16.Bg7 Rg8
17.Bd4 Bxf3 18.gxf3 Nxd4 19.cxd4 Ke7= White’s
extra pawn is fleeting.)
14...Ke7! In order to play ...f7-f6 without
blundering a pawn to Bh6-g7. Black has time to do
this because White is so under-developed.
(14...f6 15.Bg7±)
15.Bb5 f6 16.0-0 Kf7
17.Be3
(17.Rfe1 Bxc2 (17...Bxf3!? 18.gxf3 g5
19.f4 Rhg8 20.fxg5 fxg5 21.f4 Rd6 22.Bxg5 h6
156
23.h4 hxg5 24.fxg5=) 18.a4 Bd3=)
17...Nd5 18.Bc4 Bxf3 19.gxf3 Ne5 20.Bb3
Kg7 21.Bxd5 Rxd5 22.Rfd1 Rhd8 23.Rxd5 Rxd5
24.f4 Nc6=;
11.Qe2 0-0 12.0-0-0
There are a couple of novelties that Black
can try here, but I think the best practical chances
come from what has already been played —
12...exd6, because White has to find several ‘only
moves’ in order to survive. 12...exd6 Here,
innocent moves like 13.h4 (played by GM Pavlidis
in 2015) or 13.Bxd6 leave White simply worse.
a) 12...Bxc3!?N 13.bxc3 (13.dxe7 Qxe7
14.Qxe7 Bxb2+ 15.Kxb2 Nxe7 16.Nd4 Bd7
17.Bd6 Rfe8³) 13...e6! 14.Qb5! Nd5 15.Rxd5!
exd5 16.Qxd5 Be6 17.Qc5 (17.Qg5 f6 18.Qc5 Qb6
19.Qxb6 axb6³) 17...Qb6 18.Qxb6 axb6 19.Kd2
Ra4 20.Bg3 Bxa2 21.Bb5 (21.Bd3 Rd8³) 21...Ra5
22.c4 Na7 23.Re1 Nxb5 24.cxb5 Rxb5 25.Re7
Be6 26.Nd4 Rd5 27.c3 Bc8³;
b) 12...Re8N 13.dxe7 (13.Qb5 exd6
14.Rxd6 Qc7! 15.Bg3 Qe7„) 13...Qc8 14.Qb5
Bxc3 15.bxc3
 15.h4

15...Qe6! 16.Re1 Qxa2 17.Qb3 Qa1+

18.Kd2 Qa5 19.Ng5 Qd5+ 20.Qxd5 Nxd5 21.Bd6 (¹15.Bd3 Bxc3 16.bxc3 Bxd3 17.Rxd3 Qe6
Ncxe7 22.Bxe7 Rxe7 23.Rxe7 Nxe7=; 18.Kb1 Nc4„)
13.Rxd6 15...Bxc3!
a) 13.h4? Re8?! (¹13...Bxc3! 14.bxc3 Qf6 (15...h5?! 16.a4?? (¹16.Bd3²) 16...Bxc3
15.Qe3 Rfe8 16.Rxd6 Rxe3 17.Rxf6 Rxc3µ) 17.bxc3 Re4–+ 1/2 (30) –1/2 (30) Saric,I
14.Qd2 d5 15.Bd3? d4? (¹15...Nb4! 16.Bxf5 Bxc3 (2666)-Hou Yifan (2673) Wijk aan Zee NED
17.bxc3 Nxa2+ 18.Kb1 Nc4–+) 16.Nb5 Qd5 2015)
17.b3?? (17.Kb1∞) 17...Bxd3 18.Nc7 Qc5 16.bxc3 Re4 17.Be3 Qc7 18.Rd2 Qe7
19.Nxe8 Rxe8 20.Qxd3 Nb4 0–1 (20) Pavlidis, A 19.Bc5 Qf6 20.Bd4 Nxd4 21.Rxd4 Rxd4 22.Nxd4
(2522)-Zvjaginsev, V (2660) Berlin GER 2015; Bd7µ
b) 13.Bxd6? Bxc3! 14.bxc3 (14.Bxf8 Qxf8 11...0-0 12.0-0-0
15.bxc3 Qa3+–+) 14...Re8 15.Qb5 Be4 16.Bg3
Qf6µ; 12.dxe7 Nxe7 13.Bg5 Ned5!µ
13...Qc8 12...exd6 13.Rxd6

13.Bh6 Re8 14.Qd2 Bh8 15.Qxd6 Qxd6

16.Rxd6 Nb4 17.Rd2 Rec8„;
13.Bxd6 Re8 14.Bg3 Nd7 (14...Qf6!?)
15.Qd2 Qa5 16.a3 Rac8 17.Bd3 Be6µ
13...Qc8

White only has one move here — he needs

to prevent ...Nc6-b4 (a5). 14.Qb5™ 14...Re8
(14...Bg4 15.Bd3 Bxf3 16.gxf3 Nd4
17.Qb4 Ne6?! 18.Bg3 Re8 19.Rd1 Rd8 20.Ne4±
Esserman,M (2295)-Perelshteyn,E (2555)
Philadelphia USA 2008).
157
 14.Bh6 17.Qg5
(17.Qd2 Qc7 18.Bg5 Be6 19.Bf6 Nd7µ)
This option was not available to White in 17...Re6 18.Qf4
the analogous position after 11.Qe2. (18.Rd1?? Nb4!–+)
14.Qc5 This move is analogous to 14.Qb5 18...Nd8!„ (18...Re4=)
in the 11.Qe2 line: White needs to prevent 15...Bh8
...Nc6-b4(a5).
14...Re8!³ The threat is ...Bg7-f8.
(14...Bxc2 15.Kxc2 Nd4+ 16.Qxd4 Bxd4
17.Rxd4²);
14.Bb5? Nb4 15.Nd4

I think most (even strong) players would

falter and collapse here. Black threatens
...Nc6-b4(a5) and only one (highly computerized)
idea prevents it.
15...Nxa2+! 16.Kb1 16.Be3!
(16.Nxa2 Bxd4–+)
16...Nxc3+ 17.Qxc3 Qxc3 18.bxc3 Be4 16.Bb5 a6 17.Bxc6? bxc6 18.b3 Nd5!
19.f3 Bd5µ 19.Nxd5 cxd5 20.Rxd5 Be4 21.Rd6?? Qc5–+;
14...Re8 15.Qd2 16.Qf4?! Ne7! 17.Bd3 Bxc3 18.bxc3
Qxc3³;
15.Qf4!? Bxc3! 16.bxc3 Re4 16.h4?! Nb4µ
16...Qb8!?

16...Na5 This leads to a forced draw but it

158
is unnecessary to allow the ensuing chaos.
17.Bd4™
(17.Bxb6 axb6 18.Nd4 Nc6! 19.Nxf5 Qxf5
20.Kb1 Qc5µ)
17...Nac4 18.Bxc4 Nxc4 19.Qh6 Nxd6
20.Bxh8 Kxh8 21.Ng5 Kg8 22.Rd1!
(22.Qxh7+ Kf8 23.Rd1!)
22...Qc6
(22...Re2 23.Qxh7+ Kf8 24.Rxd6 Rxc2+
25.Kd1 Qc4 26.Qh8+ Ke7 27.Qe5+ Kf8=)
23.Qxh7+ Kf8 24.Qh6+ Kg8 25.g4 Bxg4
26.Rxd6 Re1+ 27.Nd1 Qe8 28.Qh7+ Kf8 29.Qh8+
Ke7 30.Qf6+ Kf8=;
16...Re6!?
By diverting White’s bishop from e3,
Black reinstates his threat of ...Nc6-b4. White is
worse if he doesn’t accept the repetition.
18.Be3 Qb8=

I had originally intended to recommend

this as the main line; perhaps some readers will
prefer this to my selection. 17.Ng5
(17.Bxb6 Bxc3 18.bxc3 axb6 19.Bc4 Rxd6
20.Qxd6 Na5 21.Bxf7+ Kxf7 22.Ng5+ Kg8
23.Qd5+ Kh8 24.Qe5+ Kg8=; 17.Rxe6 Qxe6
18.Ng5 Bxc3 19.Qxc3 Qxa2³)
17...Rxd6 18.Qxd6 Be5 19.Qc5 Bg7
20.Bc4 Nxc4 21.Qxc4 Qe8=;
16...Nb4?! (This doesn’t work here
because of a brilliant resource which was the point
of 16.Be3!) 17.Bd4! Nxc2 18.Bxh8 Kxh8 19.Bb5
Nc4 20.Qh6! Nxd6 21.Ng5 Re1+ 22.Rxe1 Qg8
23.Re7±
17.Bf4

17.Bb5 a6 18.Bxc6? Nc4!³;

17.h4?! Ne5µ
17...Qc8!

159
 CHAPTER 7 on move two (1.e4 c5 2.c3) or on move three (1.e4
c5 2.Nf3 g6 3.c3), so for convenience and
consistency I recommend a line for Black which
ANTI-SICILIANS: ALAPIN AND works against either of these. The Morra Gambit
MORRA (1.e4 c5 2.d4 cxd4 3.c3 or 1.e4 c5 2.Nf3 g6 3.d4
cxd4 4.c3), a sister to the Alapin, has recently seen
a revival, especially at the club level, thanks to the
publication of IM Esserman’s masterpiece
“Mayhem in the Morra”.
1.e4 c5 2.d4 cxd4 3.c3

1.e4 c5

Here we have the Morra Gambit, a system

Contents
1. 2.d4 cxd4 3.c3
2. 2.c3 d5 3.exd5 Qxd5 4.sidelines & 4.d4
g6 5.sidelines
3. 2.c3 d5 3.exd5 Qxd5 4.d4 g6 5.Nf3 Bg7
& 5...Nc6 6.sidelines
4. 2.c3 d5 3.exd5 Qxd5 4.d4 g6 5.Nf3 Bg7
6.Be2
5. 2.Nf3 g6 3.c3 d5
6. 2.Nf3 g6 3.c3 Nc6
1 players we offer White the option of a ‘delayed’
In these final chapters we turn to White’s
most popular anti-Sicilian options after 1.e4 c5
2.Nf3 g6: the Alapin, also known as the
c3-Sicilian, and the Morra Gambit, also known as
the Esserman Gambit. The Alapin can be played
160
emblematic of the Romantic Era in chess — not
chronologically, but in spirit: “The masters of that
time found a sound and fruitful plan: disregarding
pawns, achieve a rapid development of the pieces
for a swift attack on the enemy king. “ (World
Champion Emanuel Lasker). On Steinitzian
grounds one might insist that the principled
continuation for Black is to accept the gambit —
“the best way to refute a gambit is to accept it”.
However, there is a hidden premise in this
assertion: the gambit must be refutable! It is not at
all clear to me that the Morra Gambit is refutable;
on the contrary, Esserman’s aforementioned book
is a convincing defense of its soundness.
Furthermore, as Hyper Accelerated Dragon
Morra Gambit 1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.c3
after which our options are far more limited. As
best I can tell, Black cannot safely accept the
delayed Morra Gambit, so for the sake of
consistency I recommend declining after 1.e4 c5
2.d4 cxd4 3.c3 as well. The usual way to decline
the gambit is to transpose to the Alapin with
...Ng8-f6, but this is not in keeping with my
recommended Alapin repertoire. Fortunately, the
...d4-d3 line is a reliable way of transposing into
Maroczy-structures which we are familiar with
and happy to play as Black. I cannot overstate how
frustrating it is for Morra Gambit players to be
denied their Romantic fantasies and instead be
lulled into a slow, maneuvering Maroczy structure.
3...d3 4.Bxd3 g6 5.c4 Bg7 6.Nf3 d6 7.h3
Nf6

This idea will be studied in greater detail in

the main line. 11.Nxe5
(11.Bh6 Nxd3+ 12.Qxd3 Nd7 13.0-0 f6=)
11...dxe5 12.Bh6 Qc7 13.Qe2 Be6 14.0-0
Nd7=
8...0-0 9.Nc3 Nfd7!N

In what follows, Black is free to proceed

with traditional Maroczy development; however, I
am instead recommeding a new idea I came up
with while preparing this book based on a rapid
...Nf6-d7 and ...Bg7xc3.
8.0-0

8.Nc3 Nfd7! If Black wants to play my

idea, then White must not be permitted to
recapture on c3 with the queen, so Black must play
...Bxc3 before White achieves Bc1–e3 and
Qd1–d2.
(8...0-0 9.Be3 Nfd7 10.Qd2 This is of
course playable for Black but not in keeping with
our theme.)
9.Be3 Bxc3+ 10.bxc3 Ne5!
161
In a similar position, Esserman cautions
that White should not allow “the queen’s knight to
be lopped off by the swooping ...Bxc3 wrecking
the pristine Maroczy structure...Black’s dark
squared deficiencies are overshadowed by White’s
mangled formation.” Here, however, White does
not have much of a choice: Black is ready to
capture on c3 and White’s only ways to prevent it
are rather contrived.
9...Nc6 Black’s most popular move, but in
my judgment it is insufficiently sensitive to the
time factor in the position. White does, after all,
have more space, so he will go on to have a more
comfortable position (‘achieve harmony’) if Black
is unable to pose any disruptive threats. 10.Be3 White to avoid ...Bg7xc3; with this concession
Nd7 11.Qd2 Nc5 12.Bh6!² Black should be able to Black can equalize with ‘normal’ Maroczy moves.
defend, but White is playing for two results. 10...Nc5 11.Rb1 e6 12.Ne3
10.Be3 (12.Nc3 Bxc3 13.bxc3 e5=)
12...a5 13.Ng4 Nc6 14.Bc2 f5³;
I think if White wants to attempt an 10.Qe2 Bxc3 11.bxc3 Ne5
opening advantage then this is his best hope.
Allow the doubled pawns, and hope there is
enough dynamic compensation. In researching this
structure I have come to regard White’s attacking
potential as quite formidable, even though
computers tend to think that Black is equal ‘no
matter what’. I recommend immediately aiming
for simplifications with 11...Ne5! to neutralize
White’s initiative.
10.Bg5 Bxc3! 11.bxc3

Our go-to idea: the resulting structure is

always fine for Black as long as White can’t
immediately stir up some trouble.
(11...Nc5 12.Bc2 e5!? This interesting
blockading idea is playable as well. 13.Rd1 Nc6
14.Nd2 Be6 15.Nf1 Na5 16.Ne3 Rc8 17.Nd5 Qh4
18.Rb1 b6=)
12.Nxe5 dxe5=;
10.Qc2 White deters Black’s ...Bg7xc3 but
in so doing misplaces his queen. 10...Nc6
11...Ne5 Computers take some time to Threatening ...Nc6-b4. 11.a3 Nde5 12.Nxe5
appreciate this idea, but from a human perspective
it is clear that exchanging pieces benefits Black.
This is because White’s compensation for his
static pawn weaknesses must be predicated on
dynamic piece play.
(11...Nc5 This is perfectly playable for
Black as well, but as a matter of convenience I
have recommended 11...Ne5 here instead since I
recommend the same idea in the main line. 12.Bc2
Be6 13.e5 (13.Qe2 f6 14.Bh6 Re8=) 13...Bxc4
14.exd6 Qxd6 15.Qxd6 exd6 16.Rfd1 d5=)
12.Nxe5
(12.c5 Nxd3 13.Qxd3 dxc5 14.Qxd8 Rxd8
15.Bxe7 Re8 16.Bxc5 Rxe4=)
12...dxe5 13.c5 12...Nd4! 13.Qd1 dxe5=;
(13.f4 exf4 14.Rb1 Na6!=) 10.Bd2
13...Be6 14.Rb1 Nd7 15.Rxb7 Qc8!=;
10.Nd5 This is a pretty extreme way for
162
 said (paraphrasing) that the greatest joy in chess
preparation nowadays comes from finding ideas
the computer is unable to immediately appreciate.
That was certainly the case with this move. The
computer’s suggestions all enabled White to
develop a dangerous kingside attack, but I was
looking for something requiring less accuracy
from Black. The engine initially thinks that White
is += after the exchange on e5, but the human
realizes that either Black is getting into immediate
trouble or he’ll successfully develop his pieces and
be fine. If it is the latter, which it fortunately is,
then ‘+=’ is not a very meaningful assessment.
11...Nc5?! 12.Bxc5 dxc5 13.e5²;
This is a concession — the bishop would 11...Nc6 This is the computer’s first choice
prefer a more active square. Now we can equalize but I think the engine underestimates White’s
with ‘normal’ Maroczy development. 10...Nc5 attacking potential. 12.Nd4 Nce5 13.Be2 Nb6
11.Bc2 Nc6 12.a3 There are of course many ways 14.Bh6 Re8 15.f4 Nexc4 16.f5‚
to handle this position, I just give one relatively
straightforward and logical path for Black —
targeting the d4-square.
(12.Rb1 Nb4 13.a3 Nxc2 14.Qxc2 b6
15.b4 Ne6=)
12...b6 13.b4 Ne6 14.Bd3 Bb7 15.Rc1
Ne5! 16.Nxe5 dxe5 17.Be3 Nd4=;
10.Qd2 Ne5!
(10...Nc6 11.b3 Nc5 12.Bb2 f5 13.exf5
Nxd3 14.Qxd3 Bxf5 15.Qe3²)
11.Nxe5 dxe5 12.b3 Nc6 13.Bb2 b6
14.Rad1 Bb7=
10...Bxc3 11.bxc3

White’s attack is not devastating, but it’s

just not very comfortable to be Black here. The
position is reminiscent of lines in the Scaemish or
Rubinstein Nimzo Indian where Black doubles
White’s c-pawns and expends several tempos to
win the c4-pawn at the cost of allowing White an
f4-f5 attack.;
11...Na6

11...Ne5!

I recall an interview with Aronian where he

163
 This move fits with Nimzowitsch’s
philosophy of blockade: “First restrain, next
blockade, lastly destroy”. Unfortunately, the ideal
blockade is not always the most relevant aspect of
a position, and I believe Black’s first priority
should be to avoid a kingside onslaught by
liquidating with 11...Ne5! 12.Bc2 Nac5
(12...f6 13.Nd4 Nac5 14.Bh6 Re8 15.f4
Kh8 16.h4 Rg8 17.Re1 Nb6 18.e5 dxe5 19.fxe5
Nxc4 20.exf6 exf6 21.h5±)
13.Nd4 Re8
(13...b6 14.Nc6 Qe8 15.Bh6±)
14.f4 b6 15.f5 Ne5
(15...Bb7 16.Qf3 (16.fxg6 fxg6) 16...Ne5
17.Qf4 Kh8 18.fxg6 fxg6 19.Qh4 Kg8 20.Rad1²)
16.Nf3 Qc7 17.Qe1 Nxc4 18.e5∞
A complete mess. We should avoid this, as
the overzealous Morra Gambit player will be
rewarded for their impatience in positions like this
which require direct, un-nuanced play.
12.Nd4!
164
I believe this is White’s best attempt to
develop an initiative.
12.Bh6 Re8 13.Nd4 f6! 14.Be2
(14.f4?? Nf7–+)
14...Nbc6 15.Be3 Na5„;
12.Bc2 Be6 The simplest way for Black.
(12...Nxf3+ 13.Qxf3 Nc6 14.c5 dxc5
15.Bxc5 Qa5 16.Bd4∞; 12...Nxc4? 13.Bh6 Re8
14.Bb3 Ne5 15.Ng5 e6 16.f4 Nec6 17.f5+–)
13.Bh6 Nxf3+ 14.Qxf3 Re8 15.e5 Nc6
16.exd6 Qxd6 17.Rfb1 Na5=;
12.Nxe5 dxe5
White had better act quick, otherwise
Black will be better if he achieves ...Qd8-c7,
...Nb8-d7, ...b7-b6, and ...Bc8-a6. 13.Rb1
a) 13.Be2 Qc7 14.Qa4 Na6 15.Qa3 b6
 Nf6µ)
17...fxg6 18.Qxf8+! White can force a
draw here. 18...Nxf8 19.Rxf8+ Kg7 20.Rbf1 e6

16.Rab1 (16.Rfb1 Rd8 17.Rb5 Kg7

18.Rab1 Qd6 19.Qc1 Be6 20.a4 Nc5 21.Bxc5
(21.a5 Bd7³) 21...bxc5 22.Qe3 Rdc8=) 16...Qd6
17.Qxd6 (17.c5 Nxc5 18.Rfd1 Qf6 19.Bh6 Bd7! 21.R1f7+! Qxf7 22.Bh6+! Kf6 23.Bg5+
20.Bxf8 Kxf8 21.f3 Kg7= Black clearly has full Kg7 24.Bh6+ Kf6=
compensation, just look at the dark squares.) 12...f6!
17...exd6 18.Rfd1 Nc5 19.f3 Rd8 20.Rd2 Be6
21.Rbd1 Nb7=;
b) 13.f4 exf4 14.Bxf4 Qb6+ 15.Rf2 Be6
16.Rb1 Qc6 17.Bg5 Qc7 18.Bf4 Qc8 19.Qe2
Nd7³;
13...Qc7 14.f4 Nd7 15.f5 b6 16.Qf3
Threatening Be3-h6.
(16.Bh6 Rd8 17.Qf3 Nf6³ White cannot
make progress on the kingside, and Black is ready
to pick up the c4-pawn with ...Bc8-a6.)
16...Kh8!

This move is prophylaxis against 13.Bh6

17.fxg6
(17.Bh6 Rg8 White’s aggressive stance is
wholly illusory; Black is already clearly better.
18.fxg6 Rxg6 19.Qxf7 Ba6 20.Be3 Rag8 21.g4
165
Re8 14.f4?? when 14...Nf7! wins a piece. It also
bolsters the all-important e5 square.
12...Nxd3 This is the most obvious move
— Black would normally be happy to capture
White’s bishop. However, here Black is severely
underdeveloped, which allows White the time to
initiate a strong attack. 13.Qxd3 Nd7
(13...Qa5?! 14.Bh6 Rd8 15.f4 Qh5?! Black
deserves to be punished for parading around with
his queen in total neglect of his development.
16.Bg5 f6
 clear. In a practical game I would prefer Black,
because I think most White players would
mishandle the simultaneous demands of
aggression and restraint. 14.Bb3 Na6 (14...g5!?∞)
15.f4 Nf7 16.Bc2 Nc5 17.h4 e5 18.Nb5 b6 19.fxe5
Nxe5 20.Qxd6 Qxd6 21.Nxd6 Be6=;
13.f4 Nxd3 14.Qxd3 Na6 15.f5 Nc5
16.Qc2

17.c5!! fxg5 (17...d5 18.exd5 fxg5 19.Qc4

Bxh3 20.c6±) 18.Qc4+ Kh8 19.cxd6 exd6
(19...Bxh3 20.dxe7 Re8 21.f5!!±) 20.Qf7 Rg8
21.Qf6+ Rg7 22.fxg5+–)
14.Bh6 Re8 15.f4

16...Qe8! 17.Nb3 Nd7 White is well

advised to liquidate the situation before Black
achieves ...b7-b6 and cements White’s
c4-weakness.
(17...b6 18.Bh6 Rf7 19.Nxc5 bxc5
20.Rf4²)
18.c5 dxc5 19.Nxc5 Nxc5 20.Bxc5 b6
21.Bd4 Bb7=;
13.Bh6 Re8 14.f4?? Nf7–+
13...Na6 14.f4 Nf7 15.Bf3

Whatever the objective evaluation of this

position, I would not want to defend it as Black.
15...b6
(15...f6 16.Ne6 Qb6+ 17.Rf2 Nc5 18.Nxc5
Qxc5 19.f5 Kf7 20.Raf1²)
16.e5! Ba6
(16...dxe5 17.fxe5 Nxe5 18.Qe4+–)
17.e6 fxe6 18.Nxe6 Qc8 19.f5 Ne5 20.Qg3
Qxc4 21.Rae1 Rac8 22.fxg6 hxg6 23.Nf4±;
12...Nbd7 13.Be2! Nc5 14.f4 Ned7 15.Bf3
Nb6 16.Qe2²
13.Be2

13.Bc2 Kh8 The onus is on White to make

something happen here; Black’s ideas of The position is dynamically equal: White
completing development and blockading on c5 are has two bishops and a space advantage as
166
compensation for his weakened pawn structure.
15...Nc5

15...e5?! 16.Nb5 Qe7 17.f5 b6 18.Qd5 Rb8

19.Rad1 Rd8 20.Bg4!

20...Nc5 21.fxg6 hxg6 22.Bxc8 Rbxc8

23.Rf2 Kg7 24.Rdf1²
16.Nb3 Na4 17.c5 dxc5 18.Qxd8 Rxd8
19.Nxc5 Nxc5 20.Bxc5 Rd7

White has gotten rid of his c4-weakness

but the isolated c3-pawn still remains a liability.
The position should fizzle out into a draw; the
following is an illustration.
21.e5 Rc7 22.Bd4 fxe5 23.fxe5 Ng5
24.Bd5+ Be6 25.c4 Rd8 26.Be3 Bxd5 27.cxd5
Ne4 28.Rae1 Rcc8 29.Bxa7 Nd2 30.Rf4 Rxd5=

167
 2

1.e4 c5 2.c3

It is unusual for Black to be able to assert

central dominance so early in the Alapin. I do not
think he should fear ghosts in doing so: Black is
effectively calling White’s provocation in delaying
d2-d4 a bluff. 7.d3
In keeping with our theme of ...g7-g6, a) 7.Qa4?! e4 8.Nb5 Rc8 9.Nfd4 Bxe2
Black has three options: 10.Nxe2 Nf6µ;
1) 2...g6 3.d4 cxd4 4.cxd4 d5!? as b) 7.h3 Bh5 8.d3 (8.Qb3 0-0-0„; 8.0-0 e4
recommended by Alburt, Dzindzichashvili, and 9.Ne1 Bxe2 10.Qxe2 0-0-0„) 8...f6
Perelshteyn in their “Chess Openings for Black,
Explained”.
2) 2...Nf6 3.e5 Nd5 4.d4 cxd4 5.Nf3 g6!?
as in Adams-Kramnik, Dortmund 2013 (1/2–1/2).
3) My recommendation 2...d5 3.exd5 Qxd5
4.d4 g6!? which was also recommended by
Ftacnik in his “Grandmaster Repertoire” book on
the Sicilian.
2...d5 3.exd5 Qxd5 4.d4

4.Na3 The first time I saw this move was

in Nakamura-Yangyi, Gibraltar 2015 (1–0). It
worked out for White in that game, but I don’t
think Black has any problems if he responds
actively. 4...Nc6
(4...Nf6 5.Nf3 Nc6 6.Bc4 Qd8 7.0-0 a6
8.Nc2 Bg4 9.d4 e6 10.h3 Bxf3 11.Qxf3 Rc8
12.Rd1 cxd4 13.Bb3! Qc7 14.Nxd4 Nxd4 15.cxd4
Bd6 16.Bg5 Ke7 17.Re1± 1–0 (39)
Nakamura,H-Yu Yangyi Gibraltar 2015)
5.Nf3 Bg4 6.Be2 e5!
By further delaying his kingside
development, Black signals his intention to castle
queenside. 9.0-0 (9.Be3 0-0-0 10.0-0 Nge7 11.c4
Qg8! 12.Bxc5 Ng6!! 13.Bxf8 Qxf8 14.g4 Bxg4!!
15.hxg4 Nf4µ) 9...Nge7 10.Nc4 0-0-0 11.a3
(11.Ne3 Qd7 12.Nxe5 Bxe2 13.Nxd7 Bxd1
14.Nxf8 Be2 15.Ne6 Bxf1 16.Nxd8 Rxd8 17.Kxf1
Rxd3=) 11...Bg6

168

In this line the Accelerated Dragon player
gets to play the reverse of their usual role —
White is playing in hypermodern fashion by
relinquishing the center and seeking counterplay
on the flanks. From our experience and preference
for the Black side of the Maroczy, we know not to
underestimate the elastic potential of White’s
structure here, but it is clear nevertheless that
Black should be content with the result of the
opening. 12.Ne3 (12.Qa4 Bxd3 13.Bxd3 Qxd3
14.Be3 Nd5³) 12...Qg8 (12...Qd7=; 12...Qe6?!
13.Nh4²) 13.Qa4 Bxd3 14.Bxd3 Rxd3 15.b4
Black should trust the soundness of his
structure and enter the complications head on with
15...e4! 15...e4! (15...Kb8 16.b5 Nd8 17.Ne1 Rd6
18.Nc4 Rd5 19.f4 (19.Ne3 Rd6=; 19.Nc2 Qe6
20.N2e3 Rd7=) 19...Qe6 20.fxe5 fxe5 21.Nc2
(21.Nf3 Nf7=) 21...Ng6 22.Be3 Be7 23.Nb4!∞)
16.Ne1 (16.Nd2 Nd5 17.Nxe4 Qe6„) 16...Rxc3
17.b5 (17.Qd1 Qe6 18.Bd2 Rb3 19.Rc1 Kb8
20.Nc4 Nd4µ) 17...Nd4 18.Bd2 (18.Qxa7 Nd5
169
19.Qa8+ Kc7 20.Qa5+ Nb6 21.N1c2™ 21...Ne2+
22.Kh1 Rd3³) 18...Nd5!
If Black makes it this far he should be able
to figure the rest out over the board. 19.b6
(19.Nxd5 Qxd5 20.Bxc3 Ne2+ 21.Kh1 Nxc3
22.Qxa7 Bd6³; 19.N1c2 Nxc2 20.Nxc2 Rd3³;
19.Bxc3 Nxc3 20.Qxa7 Bd6µ) 19...Nxb6 20.Qe8+
Kc7 21.Bxc3 Ne2+ 22.Kh1 Nxc3 23.Rc1™
23...Ne2 24.Ra1™ (24.Rd1? Bd6 25.Qxg8
Rxg8–+) 24...Nc3=;
c) 7.0-0
This is oddly White’s most popular move
but it leads quite straightforwardly to a slight edge
for Black. 7...e4 8.Ne1 Bxe2 9.Qxe2 0-0-0 10.f3
exf3 11.Qxf3 Nf6³;
d) 7.Nc2 e4 8.Ne3 Qd7 9.Nxg4 Qxg4
10.Qa4 Nf6 11.Ne5 Qxg2 12.Rf1 Rc8 13.Bb5 Bd6
14.Nxc6
 14...0-0! 15.Na5 (15.Nxa7 Ra8„ ...Bb8
follows) 15...a6 16.Be2 b5µ 17...Nd5 will follow,
eyeing the f4-square. Black can take his time with
the attack: White’s king isn’t going anywhere.;
7...f6!N
This is our theme in this line: bolster the
center and aim to castle queenside.
(7...0-0-0 8.Nc4 f5 9.Ne3 Qd7 10.h3 Bh5
11.Nxf5 Qxf5 12.g4 Bxg4 13.hxg4 Qd7 14.Ng5²)
8.0-0 Nge7 9.h3
(9.Nc4 Be6 10.Re1 (10.a3 Qd7 11.b4 Nd5;
10.Nfd2 Qd7 11.Nb3 Nd5 12.d4 exd4 13.cxd4 b5
14.Ne3 c4 15.Nc5 Bxc5 16.dxc5 0-0„) 10...Qd7
11.a4 (11.a3 Nd5 12.Qc2 0-0-0 13.b4 Kb8 14.Qb2
Qc7 15.b5 Na5 16.a4 g5„) 11...Nd5 12.Bf1 0-0-0
(12...Be7 13.d4 exd4 14.cxd4 0-0³) 13.Qb3 Kb8
14.a5 g5! 15.a6 b6„)
9...Be6 10.d4 cxd4 11.cxd4 e4
170
12.Nb5 0-0-0 13.Nd2 a6 14.Bc4 Qd7
15.Nxe4 Bxc4! 16.Nbd6+ Qxd6 17.Nxd6+ Rxd6
18.Re1 Rxd4
If we accept Euler’s estimation of the value
of the pieces, then the material is numerically
equal (three pieces for a queen). However, in chess
(and in many spheres of life) there is a
phenomenon whereby the whole is greater than the
sum of its parts. My father used to refer to this as
‘cooperation points’: for every combination of
pieces, add a point, so two pieces are worth
3+3+1, ie. more than a rook+pawn (pawns don’t
count as pieces) . Similarly, three pieces are worth
3+3+3+1+1+1. This may be a bit of an
exaggeration, but it’s not a bad first
approximation. I suspect my father thought of this
himself (in the context of a variation in the Pirc
where three pieces outshine a Queen and two
pawns), but I also came across the idea in Lasker’s
Chess Manual.
4...g6

The first strong player to consistently play
this move was Yugoslavian Grandmaster Ognjen
Cvitan, who played it several times throughout the
1990s. Since then, the move has found a brigade of
supporters, most notably Radjabov, Karjakin,
Dreev, and Almasi. It was also played several
times by a young Magnus Carlsen.
5.dxc5
5.Be3 This was the ‘main line’ in the early
days of the opening, but it is no longer considered
a critical test for Black. 5...cxd4 6.cxd4
(6.Qxd4 Nf6 7.Na3 (7.Qxd5 Nxd5 8.Bd4
f6=) 7...Nc6 8.Qa4 Bg7 9.Nb5 Qd8!³)
6...Bg7 7.Nc3 Qd8 8.Nf3
8...Nh6!?
(8...Nf6 Black may prefer this mode of
development; it is largely a matter of taste. I think
many IMs and GMs prefer ideas with ...Ng8-h6
because they lead to positions which their lower
171
rated opponents (in Swiss-system tournaments)
may be unfamiliar with.)
9.Bc4
(9.Bb5+ Nd7 10.0-0 0-0 11.Bc4 Nf5=)
9...Nf5 10.Ne5 Nd6 11.Bb3 Nc6 12.Ba4
12...0-0! 13.Bxc6
(13.Nxc6 bxc6 14.Bxc6 Rb8 15.b3 Ba6µ)
13...bxc6 14.Nxc6 Qd7 15.Ne5 Qb7©;
5.Be2!? A tricky move first played by
Grandmaster Hort in 1992. 5...cxd4!
a) 5...Nc6?! 6.Bf3² (6.dxc5²) ;
b) 5...Qxg2?? 6.Bf3+–;
6.cxd4
(6.Nf3 Nf6 7.cxd4 Bg7 8.Nc3 Qd6=)
6...Bg7 7.Nf3
7...Nh6
(7...Nf6 I tend to prefer my knight on f6,
but as I have said, it is usually a matter of taste.
8.Nc3 Qd6=)
8.Nc3 Qd8 9.Bf4 0-0 10.0-0 Nf5 11.d5 a6
12.a4 Nd7 13.Qb3 b6 14.Rad1 Bb7 15.Rfe1 Rc8=
1/2–1/2 (29) Hort,V (2560)-Spraggett,K (2540)
San Bernardino 1992;
5.Na3 cxd4 6.Nb5 Na6 7.Qxd4 Qxd4
8.Nxd4 Nf6 9.Ngf3 Bg7 10.Bb5+ Bd7 11.0-0 Nc7
12.Bc4 Nfd5 13.Re1
13...f6! Black flexes his muscles; the center
is his for the taking. 14.a4 a6=
5...Qxd1+ 6.Kxd1
6...Nf6
Black needn’t rush to win back his
sacrificed pawn. Sometimes he will even play
...b7-b6, abandoning hopes to regain the pawn in
favor of opening up lines for his pieces. Even with
queens off the board, White’s king is not safe in
the center.
6...e6N I toyed around with this new idea
for a while, but I came to the conclusion that it is
unnecessary since the main move, 6...Nf6,
equalizes comfortably. I have included my
172
analysis of 6...e6 in case the reader is interested.
7.b4!
a) 7.Be3 Nf6
8.Nf3
a1) 8.Bd4 Be7 9.Nf3 0-0 10.Nbd2 Rd8
11.Kc2 Nd5 12.Bc4 Nc6 13.Rhe1 Nxd4+ 14.cxd4
b6 15.Bxd5 Rxd5 16.b4 Bb7©;
a2) 8.Na3 Nbd7 9.Nb5 (9.Bb5 Ng4 10.c6
bxc6 11.Bxc6 Rb8©) 9...Nd5 10.Nf3 Nxe3+
11.fxe3 Ke7 12.b4 a5©;
8...Be7 9.Na3 0-0 10.Nb5 Nd5 11.Kc2
Nxe3+ 12.fxe3 a5 13.Nd6 Nd7 14.b4 Nf6 15.Nxc8
Rfxc8 16.Kb3 b6 17.Bc4 (17.cxb6 Ne4 18.b7
Rxc3+ 19.Kb2 Rb8–+) 17...bxc5 18.b5 Ne4„;
b) 7.Bb5+ Bd7 8.Bxd7+ Nxd7 9.b4 Ne7
10.Kc2 Bg7 11.Nd2 0-0 12.Ne4 Nd5 13.Bb2 b6
14.c6 (14.cxb6 N7xb6³) 14...Ne5 15.b5 a6 16.a4
Nc4 17.Ne2 Nc7 18.bxa6 Rfc8µ;
7...Bg7 8.Kc2
8...Ne7
(8...a5 9.Bb2 (9.b5 Nd7 10.Ba3 Ngf6
11.Bd3 Bf8 12.c6 bxc6 13.bxc6 Nc5 14.Bb5 e5=) 1/2–1/2 (28) Pedersen,N (2508)-Nevednichy,V
9...Nf6 10.Nd2 0-0 11.Ngf3 axb4 12.cxb4 Bd7 (2569) Istanbul 2003)
13.Bc4 Bc6 14.Rhg1 Rd8 15.Bb3 Bb5 16.Nc4 Nc6 8...e6!
17.a3 Ne4 18.Bxg7 Kxg7 19.Rge1±)
9.Nf3 0-0 10.Bf4
(10.Nbd2 Nd5 11.Bb2 b6 12.Ne4 (12.cxb6
axb6©) 12...bxc5 13.Nxc5 Nd7 14.Ne4 Bb7
15.Nd6 Rab8 16.a3 Bc6 17.b5 Ba8 18.c4™
18...Bxb2 19.Kxb2 Nf4 20.Kc3 Rb6 21.Rd1 Nc5
22.Kb4 Na6+ 23.Ka5 Nc5 24.Kb4=)
10...Nd5
(10...Nec6 11.b5 e5 12.Bg3 (12.Be3 Ne7=)
12...Bf5+ 13.Kb2 Na5 14.Nbd2 (14.Bxe5 Bxe5
15.Nxe5 Rd8©) 14...Nd7 15.Nb3 Nxb3 16.axb3
e4 (16...Nxc5 17.Bxe5 Bxe5 18.Nxe5 Rfd8 19.Bc4
Rd2+ 20.Ka3±) 17.Nd4 Nxc5 18.Bd6 Rfc8
19.Nxf5 gxf5 20.Bxc5 Rxc5 21.Bc4²)
11.Be5 a5 12.Bxg7 Kxg7 13.Kb3 Nc6 (8...Bg7 9.Nf3 0-0 10.Ke2 Rc8 11.Be3
Nd5 12.Rhd1 Nxe3 13.Kxe3 Bxb5 14.Nxb5²
1/2–1/2 (57) Madeira,W (2215)-Leitao,R (2566)
Sao Paulo 2004)
9.Bxd7+ Nbxd7 10.Nb5
(10.b4 Ne4³)
10...Ke7! 11.b4 a6 12.Nd6 b6³
7...Bg7 8.Bb5+

8.Be3 Ng4 9.Bb5+ Bd7 10.Bxd7+ Nxd7

11.Re1 Nxe3+ 12.Rxe3 Nxc5 13.Kc2 e6 14.Nbd2
Ke7 15.Nb3 Rac8= 0–1 (49) Ragger, M
(2265)-Kozul,Z (2587) Maribor 2003.
8...Bd7

14.Bc4
(14.b5 Nb8!=)
14...axb4 15.Bxd5 Na5+
(15...exd5 16.cxb4 Re8 17.b5 Na5+ 18.Kc3
Bd7 19.Nd4²)
16.Kc2 exd5 17.cxb4 Bf5+ 18.Kc3
(18.Kb2!±)
18...Nc6 19.a4 d4+ 20.Nxd4 Nxd4
21.Kxd4 Rfd8+ 22.Kc4 b5+ 23.cxb6 Rac8+
24.Kb3 Be6+ 25.Kb2 Rd4 26.Ra3 Rxb4+ 27.Ka1
Rxb6=
7.Nf3

7.Bb5+ Bd7 8.Na3

(8.Bxd7+ Nbxd7 9.c6 bxc6 10.Nf3 e5 9.Bd3
11.Re1 Ng4 12.Kc2 f6 13.Be3 Nxe3+ 14.Rxe3
Bh6 15.Re2 Rd8 16.Nbd2 Bxd2 17.Nxd2 Kf7= 9.Bxd7+ Nbxd7 10.Kc2 Nxc5 11.Be3 Rc8
173
12.Re1 Nd5 13.Bd4 0-0 14.Bxg7 Kxg7 15.Nbd2
b5 16.g3 Rfd8 17.Rad1 e6 18.Nb3 Na4 19.Rd4
Naxc3 20.Kd2 Nxa2 0–1 (21) Sriram,J
(2356)-Kunte,A (2487) Mumbai 2000.
9...Bc6 10.Be3 Ng4 11.Nbd2 Nd7 12.Nb3
0-0-0 13.Kc2 Nde5 14.Nxe5 Nxe5 15.Bf1 Be4+
16.Kc1 Ng4³

1/2–1/2 (53) Smerdon,D (2470)-Saric,I

(2559) Dresden 2008.

174
 3

1.e4 c5 2.c3 d5 3.exd5 Qxd5 4.d4 g6

5.Nf3

I think this rare move deserves more

attention.
(6.Na3 This transposes to subchapter 5,
1.e4 c5 2.Nf3 g6 3.c3 d5 4.exd5 Qxd5 5.Na3 Bg7
6.d4.)
6...cxd4
5...Nc6!? (6...Qe6+?! We will see this is the right
move after 5...Nc6 6.Nbd2 but not here. 7.Be2
In this book I have tried, where possible, to cxd4 8.Nxd4 Bxd4 This is the reason: instead of
present novel ideas, even where the accepted trading on d4 with a knight as in the 5...Nc6 line,
wisdom is perfectly adequate. There is value in Black has to cede the two bishops to White. 9.cxd4
exploring new terrain, both from a scientific Nf6 10.0-0 0-0 11.Bf3²)
perspective of enriching the literature on an 7.Bc4 Qd8
opening, but also from the practical, ‘surprise (7...Qh5N
value’ perspective. 5...Nc6 is an extremely rare
move, about fifty(!) times less common than the
main move, 5...Bg7. I am not sure, nor do I frankly
care very much, which of 5...Bg7 or 5...Nc6 is
‘objectively’ the better move. It seems to me
though, that the new ideas which follow lead to
rich, unexplored positions where a well-prepared
Black player can pose serious problems for their
opponent. Praxis and the test of time will
determine whether these ideas have wings, or
whether Black is better off adhering to the usual
5...Bg7. Before continuing I would like to mention
that the following is joint analysis between myself
and Canadian (exiled by marriage to America) IM
Dave Ross. Black does not quite equalize here. 8.0-0!
5...Bg7 The reader may consult GM (8.Qb3 Nh6 9.Nxd4 0-0 10.N2f3 Nf5 11.0-0 Nd6
Ftacnik’s book on the Sicilian for his treatment of 12.Bd3 Nd7=) 8...Nc6 (8...dxc3? 9.Ne4! cxb2?
this move. Black may be fine with accurate play, 10.Bxf7+! Kxf7 11.Qb3+ Kf8 12.Neg5 Nh6
but the amount of awkward queen moves Black is 13.Bxb2 Bxb2 14.Qxb2 Rg8 15.Nxh7+ Ke8
lured into is not to my taste. 6.Nbd2! 16.Nfg5+–) 9.Qb3 (9.cxd4 Nf6 10.d5 Nxd5
11.Qb3 e6 12.Ne4 0-0 13.Ng3 Na5! 14.Qb5 Nc7!=
175
Alas, Black hangs on by the skin of his teeth.)
9...Nh6 10.Nxd4 Nxd4 11.cxd4 0-0
12.h3! A strong prophylactic move: Black
can no longer disentangle with ...Nh6-f5 in light of
g2-g4. 12...Bxd4 (12...g5?! 13.g4 Nxg4 14.Be2
(14.hxg4 Bxg4 15.Kg2 Bf5=) 14...Qh4 15.hxg4
Bxg4 16.Nf3±) 13.Re1 Nf5 (13...Bg7 14.Nf3 Nf5
15.Bf4²) 14.Ne4 Nd6 15.Nxd6 exd6 16.Re7 g5
17.Bd2! g4 18.hxg4 Bxf2+ 19.Kf1! (19.Kxf2
Qh4+ 20.Kg1 Qxe7=) 19...Qg6 20.Kxf2 Qf6+
21.Qf3 Qxe7 22.Rh1±)
8.Nxd4 Nh6
(8...Bxd4 9.cxd4 Qxd4 10.0-0 Nc6
11.Qb3²; 8...Nf6 9.0-0 0-0 10.N2f3²)
9.N2f3 0-0 10.0-0
Black’s position is not pretty; I prefer the
fresh positions which arise after 5...Nc6!?.
10...Nf5 11.Re1
(11.Bg5 Qc7 12.Qe2²)
11...Nxd4 12.Nxd4 Qc7 13.Bb3 a6 14.Bg5²
1/2–1/2 (57) Onischuk,A (2664)-Kudrin,S (2549)
176
Tulsa 2008.
6.Be3
The main move 6.Be2 will be covered in
the next subchapter. 6.dxc5 This is premature; if
White wants to take on c5 then he should first play
6.Be3 then take on c5 after Black commits his
bishop to g7. 6...Qxd1+ 7.Kxd1 Nf6 8.Be3 Nd5
9.Nbd2 e5!
10.Ne4 Bf5 11.Bd3 0-0-0 12.Ke2 Nxe3
13.fxe3 Be7 14.Nd6+
(14.Bc2 Bg4 15.Nf2 Bxf3+ 16.gxf3
Bxc5=)
14...Bxd6 15.Bxf5+ gxf5 16.cxd6 Rxd6
17.Rad1 Kd7!=;
6.Na3
One of the nice things about this 5...Nc6
system is that it is particularly well-suited to meet
6.Na3, which is the most popular move against
5...Bg7. Indeed, when I was practicing this line in
online blitz I came across Alapin specialist GM A.
Zhigalko and 6.Na3 was his choice as well. As we
see, though, Black’s 5...Nc6 enables him to
quickly castle queenside and pose White with
immediate problems. 6...Bg4 7.Nb5
(7.Be2 cxd4 8.cxd4 Nf6 9.Nb5 Rc8 10.Nc3
Qa5 11.d5 Rd8 12.0-0 Bg7³; 7.Bc4 Bxf3 8.gxf3
Qf5„ It is a great practical upshot of this system
that natural moves by White allow Black to
equalize with ease. 9.Bd3 Qh5 10.Be4 Nf6
11.Nb5? Nxe4 12.Nc7+ Kd7 13.Nxa8 Nf6µ)
7...0-0-0!

Black gets to have some fun in this line.

8.Na3
(8.Nc3 Nc2+ 9.Kd2 Nxe3 10.fxe3 Qc6
11.Ne5 Qa6=)
8...Bh6! 9.Nd2 Qg4 10.Nb5 Bf5!

White is suddenly caught off guard —

Black isn’t supposed to be able to castle queenside
in this opening! Reap the rewards of 5...Nc6.
Black has taken over the initiative and White
needs to be extremely accurate to not lose outright
in the next few moves. 8.Be2™
(8.Be3 a6 9.a4 Bxf3 10.gxf3 Bh6!µ)
8...a6 9.a4
(¹9.Na3 cxd4 10.cxd4 Bxf3 11.Bxf3 Qxd4
12.Qc2 Qd3 13.Qxd3 Rxd3 14.Nc4©)
9...cxd4 10.cxd4 axb5 11.axb5 Bxf3
12.bxc6 Bxe2 13.Ra8+ Kc7 14.Bf4+ e5 15.Bxe5+
Qxe5–+;
6.c4?! Qe4+ 7.Be3 Nb4!
If you ever get this position over the board,
relish your opportunity — your opponent is likely
a masochist, and you may get to enact one of the
beautiful variations which follow. 11.f3
(11.Nc7+ Kd8 12.f3 Qh4+ 13.Bf2 Qf6
14.Nxa8 Nc2+ 15.Ke2 Qe6+ 16.Ne4 Bxe4
17.dxc5+ Kc8–+)
11...Qh4+ 12.Bf2
(12.g3 Qf6 13.Bxh6 Nc2+ 14.Kf2 Nxh6
15.Rc1 a6 16.Nc7+ Kd8µ)
12...Qg5

177
 13.Rc1
(13.Nc7+ Kf8 14.h4 (14.Nxa8 Nc2+ 6...Bh6 7.dxc5 Qxd1+ 8.Kxd1 Bxe3
15.Ke2 cxd4–+) 14...Qf4 15.Nd5 Nxd5 16.cxd5 9.fxe3²;
cxd4µ) 6...Nf6 7.c4 Qd6 8.d5 Ne5 9.Be2²
13...Qf4 14.g3 Qb8 15.a3 Na2 16.Ra1 a6 7.dxc5

7.Na3 Black is fine after the mass

exchanges on d4. 7...cxd4 8.Nb5 Qd8 9.Nbxd4
Nxd4 10.Bxd4 Bxd4
(10...Nf6!? 11.Bb5+ Bd7 12.Qe2 0-0
13.Rd1 Bxb5 14.Qxb5 Qc8=)
11.Qxd4 Qxd4 12.Nxd4 Nf6

17.Rxa2 axb5 18.Qb3

(18.cxb5 Nf6 19.Bg2 0-0 20.0-0 cxd4µ)
18...cxd4 19.Bxd4 bxc4 20.Nxc4
(20.Bxc4 Qd6!=)
20...Nf6 21.Be5 Qd8 22.Qb5+ Bd7
23.Qxb7 0-0©;
6.Nbd2N Since this was a big problem in
the 5...Bg7 line I searched for games after 5...Nc6 13.0-0-0 Bd7 14.Be2 0-0-0 15.Rhe1 e6
6.Nbd2. Turns out it is a novelty, but not one 16.Bf3 g5!=
Black needs to worry about. 6...Qe6+! 7.Be2 cxd4 7...Qxd1+ 8.Kxd1 Nf6
8.Nxd4
(8.cxd4 Nf6 9.0-0 Bg7 10.Nc4 0-0 11.b3
Nd5 12.Bb2 Rd8=)
8...Nxd4 9.cxd4 Nh6 10.0-0 Bg7 11.Bb5+
Bd7 12.Re1 Qg4!=
6...Bg7
178
 Black is down a pawn, and will likely stay
down a pawn for the foreseeable future, but as
compensation he has better control over the center
and the potential to harrass White’s king which
has lost its castling rights.
9.Bc4
9.Bd3 Nd5 10.Nbd2
(10.Be4 0-0! 11.Bxd5 Rd8 12.Ke2 Rxd5
13.Na3 b6! 14.Nb5 (14.cxb6 Ba6+ 15.Ke1 axb6
16.Rd1 Ra5! 17.Nd4 Ne5„) 14...Rb8 15.Nc7 Rd7
16.cxb6 axb6 17.Nb5 Ba6 18.a4 Na7=)
10...0-0 11.Ke2 Rd8 12.Rhc1 h6©
Prophylactically guarding the g5-square. Black
will continue ...e7-e5/...Bc8-Be6/...f7-f5 etc. White
can only hope he is resilient enough to endure this,
because he has no aggressive ambitions of his
own.
9...0-0 10.Nbd2 Rd8 11.Ke2 Na5 12.Bd3
12.Rhd1 Nxc4 13.Nxc4 Nd5=
12...Nd5
179
13.Nb3
13.Nc4 Bf5! 14.Nxa5 Nxc3+ 15.bxc3
Bxd3+ 16.Kd2 Be4+ 17.Bd4 e5=;
13.Rac1 b6! 14.b4
(14.cxb6 axb6 15.a3 Nb7!„)
14...Nb7 15.c6 Nd6 16.Nd4 e5 17.Nb5
Bg4+ 18.f3 Bf5 19.c7 Bxd3+ 20.Kxd3 Rd7
21.Nxd6 Rxd6³
13...Nc6
Black is happy to ‘just play’ this position
down a pawn, without feeling any urgency to
regain it. Petrosian once said (paraphrasing) that
‘too often players sacrifice a pawn and then go on
to play as if they blundered it’.
14.Rhd1
14.Rac1 e5 15.Rhd1 h6 16.Nfd2 f5 17.f3
Be6 18.Nc4
18...e4! 19.fxe4 fxe4 20.Bxe4 Nxc3+
21.bxc3 Bxc4+ 22.Kf2 Rf8+ 23.Kg1 Rae8=;
14.Bc2 a5 15.a4 Be6=
14...Bxc3!? 15.bxc3 Nxc3+

16.Kf1

16.Ke1 Nb4 17.Be2 Nc2+ 18.Kf1 Nxd1

19.Rxd1 Rxd1+ 20.Bxd1 Nxe3+ 21.fxe3 Be6
22.Nfd4 Bd5 23.Bf3 Bxf3 24.gxf3 b6= A rook is
no worse than two knights in the endgame.
16...Nxd1 17.Rxd1 Nb4 18.Nbd4 Nxd3
19.Rxd3 Be6 20.Rd2 Bc4+ 21.Ke1 Bd5=

It may not be clear from the looks of this

position, but it will be clear that Black is no worse,
perhaps even to be preferred, once he achieves
...f7-f6 and ...e7-e5.

180
 4 0-0
1.e4 c5 2.c3 d5 3.exd5 Qxd5 4.d4 g6
5.Nf3 Nc6!? 6.Be2 Bg7
7.c4
7.dxc5 We do not take on d1 this time
because White can retain his castling rights by
recapturing with the bishop. 7...Qxc5
8.Na3
(8.Be3 Qa5 9.0-0 Nf6 10.Na3 0-0 11.Nc4
Qc7 12.Qc1 Nd5 13.Bh6 e5=)
8...Nf6 9.Nb5 0-0 10.Be3 Qf5=;
7.Na3 cxd4 8.Nb5
(8.cxd4 Nf6 9.Nb5 0-0³)
8...Qd8! 9.Nfxd4 Nxd4 10.Nxd4 Nf6
11.Qa4+ Bd7 12.Qb4 Qb6 13.Qxb6 axb6 14.a3
0-0 15.0-0 Nd5 16.Bc4 Bc6 17.Rd1 e6=;
7.Be3 cxd4 8.cxd4 Nf6 9.Nc3 Qa5 10.0-0
181
This is a well-known position from the
Tarrasch Queen’s Gambit Declined (Rubinstein
Variation), with colors reversed. Thus, White is
effectively playing the Tarrasch QGD a tempo up
here, but this tempo is only enough to allow him to
equalize, and in fact I still prefer Black. 11.Qb3
(11.a3 Be6 12.b4 Qd8=)
11...Qb4! 12.d5 Qxb3 13.axb3 Nb4 14.Bc4
14...b5! 15.Nxb5
(15.Bc5 bxc4 16.Bxb4 cxb3 17.Bxe7 Re8
18.d6 Be6)
15...Nfxd5 16.Rxa7 Nxe3 17.Rxa8 Nxf1
18.Kxf1 Ba6 19.Rxf8+ Kxf8 20.Nc3 Bxc4+
21.bxc4 Nd3=
7...Qd8!
The queen turns out to be happier on d8
than d6 because it won’t come under attack from
White’s Nb1–a3-b5 or Nf3-g5-e4.
7...Qd6 8.d5 Nd4
 (8...Ne5 9.Ng5! (9.Nxe5 Bxe5 10.Nc3 Nf6
11.Bh6 Bd7 12.h3 0-0-0=) 9...Nh6

10.Ne4 (10.Nc3 Nf5 11.Nge4 Qc7=)
10...Qb6 11.Nbc3 Nf5 12.0-0 0-0 13.b3 Nd4
(13...Qd8 14.Bb2 b6 15.Qd2 Bb7 16.Rae1²)
14.Bb2 Bd7 (14...f5 15.Na4±) 15.f4 f5 16.fxe5
fxe4 17.e6±)
9.Nxd4!
(9.Nc3 Nxf3+ 10.Bxf3 Nh6 11.0-0 Nf5„)
9...cxd4 10.0-0 Nf6 11.Na3!
9.Nc3
9.Nxd4 cxd4 10.0-0
(10.Nd2 e5! (10...e6!? 11.Ne4 exd5
12.cxd5 Ne7 13.Bb5+ Kf8∞) 11.0-0 (11.f4 exf4
12.0-0 Ne7=) 11...Ne7 12.Re1 0-0 The position is
rich, and offers mutual chances. 13.Bf1 b6 14.a4
Nf5 15.a5

15...b5! 16.b3 Re8 17.Bd3 Rb8 18.Ba3

11...0-0 12.Nb5 Qb6 13.a4 Ne4 14.Bf3 Nd6=)
(14.a5 Qf6 15.Bf3 Nd6 16.Nxd6 exd6 10...e5!
17.b4 Bf5 18.c5 d3 19.Ra3 Qd4„)
14...Nd6 15.Nxd6 exd6 16.b3²
(16.Be4²)
8.d5

8.dxc5 Qxd1+ 9.Bxd1 Be6 10.Be2 Nh6

11.Nc3 Nf5 12.0-0 0-0 13.Rd1 Rfd8 14.Bf4 Nfd4
15.Nxd4 Nxd4 16.Bd3 Rac8=
8...Nd4
182
 11.Bxh6 (The concept for this move is
borrowed from a well known line in the English:
1.c4 c5 2.Nf3 Nc6 3.g3 g6 4.Bg2 Bg7 5.Nc3 d6
6.0-0 Nh6 7.d4! cxd4 8.Bxh6! Bxh6 9.Nxd4).
11...Bxh6 12.0-0 0-0 13.Re1 Bd7 14.Qe2
(14.Qb3 Rb8 15.Qa3 b6=)
14...Re8 15.Bg4

It is very unclear which side’s passed

pawns are stronger here, but determining this over
the board will certainly lead to an interesting fight.
(10...e6!? 11.Re1 (11.dxe6 Bxe6 12.Qa4+
Qd7 13.Qxd7+ Bxd7=) 11...Ne7 12.Qa4+ Bd7
13.Qa3 0-0 14.Bg5 f6 15.Bd2 exd5 16.Ba5 b6
17.Bb4 Rf7 18.Bf3 Bf8∞)
11.Re1 15...f5! 16.Bf3 a6 17.a4 Rb8 18.a5 Bg7=
(11.dxe6 Bxe6 12.Qa4+ Qd7 13.Qxd7+ 11...Nf5 12.Re1 0-0 13.Bg5
Bxd7 14.Re1 0-0-0=; 11.Nd2 Ne7 12.Qa4+ Bd7
13.Qb3 b6 14.f4 exf4 15.Ne4 0-0 16.Bxf4 Nf5=) 13.Bf4
11...Ne7 12.Bd3 0-0 13.a4 b6
(13...f5!? 14.f3 f4 15.Nd2 Nf5„)
14.b4 Bf5 15.Bf1 Re8=
9...Nxf3+ 10.Bxf3 Nh6

White may do without provoking ...h7-h6,

but Black continues in more or less the same way.
13...Nd4 14.Be4
(14.Rc1 a6 15.b3 Re8 (15...Bd7?! 16.Bg5
Re8? 17.d6±) 16.Be4 Bd7 17.Qd2 f5 18.Bd3 e5
Once Black plants his knight on d4 he 19.dxe6 Nxe6=)
should be well positioned for anything White 14...Re8 15.Rc1 Bd7 16.Qd2 f5 17.Bd3 e5
throws at him. 18.dxe6 Nxe6 19.Be5 Bc6=
11.0-0 13...h6 14.Bf4 Nd4

183
 As in the Breyer Variation of the Maroczy, 31...f3!! 32.gxf3 Qf4 33.Qb7 Rab8
Black’s knight on d4 is a nuisance for White to 34.Nxe6 Rxb7 35.Nxf4 Rxe1+ 36.Rxe1 gxf4=
play around. The following variations illustrate 22...Bxd4
some possible ways the game might continue, but
in general Black can probably just figure things
out over the board from here. The position is
roughly equal.
15.Rc1

15.Be4 g5 16.Be3 e5 17.dxe6 Bxe6„

15...Bd7 16.Be4 Re8 17.Qd2 g5 18.Be3 f5
19.Bd3 e5!

23.b4 b6 24.bxc5 bxc5 25.Rxe6 Qxe6

26.Nc7 Qf6 27.Nxa8 Rxa8=

20.dxe6 Bxe6 21.Nd5 Qd7 22.Bxd4

22.b4 b6 23.bxc5 bxc5 24.Qa5 f4 25.Bxd4

Bxd4 26.Nc7 Qd8 27.Qa6 Bxf2+ 28.Kxf2 Qd4+
29.Kf1 Qxd3+ 30.Kg1 Qd4+ 31.Kh1

184
 5

1.e4 c5 2.Nf3 g6 3.c3

7...Qxe5 8.Qxb7 Qxe4+ 9.Kf1

(9.Kd1?? Qa4+ 10.b3 Qc6–+)
9...Qd3+ 10.Ke1
(10.Kg1? e6 11.Qxa8 Bd6µ)
10...Qe4+ 11.Kf1=
4...Qxd5 5.Na3
This move order is recommended by GM
Sveshnikov in his book ‘The Complete c3 Sicilian’
as particularly effective against the system I am
recommending against 2.c3. White’s idea is to
delay d2-d4, so that after 3...d5 4.exd5 Qxd5
White has the option of 5.Na3, considered better
for White by Sveshnikov. Black has two ways to
transpose to the main line of this chapter: either
play 3...d5 and allow Sveshnikov’s 5.Na3, or play
3...Nc6 and allow a ‘delayed’ Rossolimo (4.Bb5).
According to my analysis, either of these is fine
for Black, but 3...Nc6 is more in keeping with my
recommended repertoire for reasons I will make
clear in this subchapter.
3...d5 4.exd5
Sveshnikov is quite optimistic about
4.Bb5+ This effectively leads to a forced White’s chances: “White develops quickly and
draw, which may be (for some) another reason to takes a firm grip on the initiative”. 5.d4 Nc6 This
prefer 3...Nc6. 4...Bd7 5.Bxd7+ Qxd7 6.Ne5 Qe6 transposes to the main line of the Alapin (2.c3 d5
7.Qb3!? 3.exd5 Qxd5 4.d4 g6 5.Nf3 Nc6).
5...Bg7 6.d4

This transposes to 1.e4 c5 2.c3 d5 3.exd5

Qxd5 4.d4 g6 5.Nf3 Bg7 6.Na3, and I think it is
fine for Black. However, I recommend (in the
main line of the Alapin section) that Black plays
5...Nc6 instead of 5...Bg7, so this variation does
not ‘come up’ in the main move order. That being

185
said, I recommend 5...Nc6 because I don’t like (12.Be2 Rd8=) 12...Rd8 13.Qe2 Nd5 14.d4 cxd4
Black’s position after 5...Bg7 6.Nbd2! which is not 15.Ncxd4 Nxd4 16.cxd4 Bf6!= (16...e6? 17.Bg5²);
the case here, so some readers may prefer to allow b) 9.Nb5 0-0 10.Re1 Rd8 11.d4 Bg4! 12.h3
this line rather than allowing a delayed Rossolimo Bh5!
with 3...Nc6.
6.Bc4 This is Sveshnikov’s
recommendation (following up with: 6...Qe4+
7.Be2). 6...Qe4+ 7.Be2
a) 7.Qe2 Qxe2+ 8.Bxe2 (8.Kxe2 a6!=)
8...Nf6 9.d4 (9.Nb5 Nd5=) 9...cxd4 10.Nxd4 0-0
11.0-0 Rd8 12.Rd1 Bd7 13.Bf3 Nc6=;
b) 7.Kf1 Be6 8.Be2 Nc6 9.Ng5 Qf5
10.Nxe6 Qxe6 11.Qb3 Qd7=;
7...Nf6 8.0-0

13.Be3 (13.g4? Bxg4 14.hxg4 Qxg4+

15.Kf1 Qh3+ 16.Kg1 cxd4 17.cxd4 Rd5µ)
13...Nd5 14.Qc1 Nxe3 15.Qxe3 Qxe3 16.fxe3
Bxf3 17.Bxf3 cxd4 18.exd4 e5 19.dxe5 Nxe5=;
9...cxd4 10.Nb5
(10.Nxd4 Nxd4 11.cxd4 0-0 12.Bf3 Qf5
13.Re1 Nd5=)
10...0-0 11.Nfxd4 Nxd4 12.Nxd4 e5

We have been following Sveshnikov’s

recommendations from his book, but here he only
gives 8...0-0 which Miezis played against him in
2005. The improvement for Black was first played
by Van Kampen in 2011.
a) 8.Nb5 Qc6 9.d4 (9.Ne5 Qb6 10.Nc4
Qd8 11.d4 a6 12.Nba3 b5 13.Bf3 Ra7 14.Bf4
Nfd7=) 9...0-0 10.0-0 a6 11.Ne5 Qb6 12.dxc5
(12.Na3 Be6=) 12...Qxc5 13.Qd4 Nbd7=;
b) 8.d4 cxd4 9.Nb5 (9.Nxd4 Qxg2 10.Bf3
Qh3µ; 9.Qxd4 Qxd4 10.Nxd4 0-0 11.0-0 Rd8
12.Rd1 Bd7 13.Bf3 Nc6=) 9...0-0 10.Nbxd4
(10.Nc7 dxc3 11.Nxa8 Nd5–+; 10.Qxd4 Qxd4 13.Nb5 Qc6 14.Qd6 Bd7 15.Rd1 a6
11.Nbxd4 Bd7 12.0-0 Nc6 13.Nb3 Rad8 14.Nc5 16.Bg5 Rad8 17.Qxc6 Bxc6 18.Nd6 h6 19.Bh4
Bc8=) 10...Nc6 11.0-0 Bg4=; Rd7 20.Nc4 Nd5 21.Bg3 Re8 22.a4 Nf4³ 1/2–1/2
8...Nc6! (40) Michalczak,T (2325)-Van Kampen,R (2519)
(8...0-0 9.d4 cxd4 10.Nxd4² Dortmund 2011.
Sveshnikov-Miezis 2005) 6...cxd4
9.d4
a) 9.Re1 0-0 10.Bc4 (10.Bb5 Qd5 11.Bxc6
Qxc6 12.Rxe7? Be6µ) 10...Qf5 11.Nc2 Qh5 12.h3
186
 12.h3 Qh5 13.Qd2
(13.0-0 Nf5=)
13...Ng4 14.Bg5 h6 15.Bh4 Nf6=
9...0-0 10.0-0 Nf5 11.Qb3 e6

7.Bc4

7.Nb5 Na6 8.Nbxd4 Nf6 9.Bb5+ Bd7

10.Qe2 0-0 11.Bxa6 bxa6= 12.Qxa6?? e5 13.Ne2
Bb5 14.Qa5 Qd3 15.Ng3 Nh5–+
7...Qe4+ 8.Be3 Nh6
Not 8...dxe3?? 9.Bxf7+ winning.
9.cxd4
9.Nb5 0-0 10.cxd4 Nc6 11.Nc3
I’ve seen some grandmasters in America
play this line against lower rated players in Swiss
tournaments presumably to obtain winning
chances but in my view White is playing with little
risk: the isolated pawn is inadequately blockaded,
so it can always be pushed if White is content to
neutralize the position.
12.Rae1

12.Nb5 Nc6 13.Nc3

(13.Bd3 Qd5 14.Qa3 Nxe3 15.fxe3 Qd8=)
13...Qg4 14.Be2 Nxe3 15.fxe3 Qh5=
12...Nxe3 13.Rxe3

13.fxe3 Nc6=
13...Qf4 14.d5 exd5 15.Bxd5 Na6 16.Rc1
Qf6=

Black has two equally viable options here.

11...Qf5 is riskier, but perhaps more fun as it
involves an unclear piece sacrifice. 11...Qg4!?
(11...Qf5!? 12.h3 (12.0-0 Ng4 13.h3 Nxe3
14.fxe3 e6= Black will be fine after he relocates
his queen to e7 via f6.) 12...Qa5 13.g4 (13.d5 Ne5
14.Nxe5 Bxe5 15.Bxh6 Bxc3+ 16.Kf1 Bg7
17.Bxg7 Kxg7 18.Qd4+ f6 19.Re1 Qb4=)
13...Nxg4! 14.hxg4 Bxg4©)

187
188
 6 whatsoever. 5...dxe4 6.Qxd8+ Nxd8 7.Nd4 Bg7
8.Bb5+ Bd7 9.Bxd7+ Kxd7 10.0-0 Nc6 11.Rd1
Nxd4 12.cxd4 f5 13.Nc3 Ke6! 14.d5+ Kf7=;
1.e4 c5 2.Nf3 g6 3.c3 Nc6 5.exd5 Qxd5 This transposes to 1.e4 c5 2.c3 d5
3.exd5 Qxd5 4.d4 g6 5.Nf3 Nc6.)
5...cxd4 6.cxd4 Bg4 Black has ideas of
...Qd8-b6, ...Ng8-h6-f5, etc. The position plays
itself. 7.Be2
a) 7.Nbd2 Nh6 8.h3 Bxf3 9.Nxf3 Nf5
10.Qa4 (10.g4 Ng7=) 10...a6 11.Bd3 Qd7 12.Bxf5
Qxf5 13.Qb3 Qd7=;
b) 7.Bb5 Qb6 8.Bxc6+ Qxc6 9.0-0 e6
10.h3 Bxf3 11.Qxf3 Bg7 12.Nc3 Ne7 13.Bg5 Nf5
14.Ne2 h6 15.Bd2 0-0=;
7...Nh6 8.Nbd2

I prefer this move order, because I quite

like the delayed Rossolimo (4.Bb5) for Black,
especially if it is an Alapin player on the White
side.
4.Bb5

This transposes to a harmless variation of

the Rossolimo: 1.e4 c5 2.Nf3 Nc6 3.Bb5 g6 4.c3.
4.d4 d5 5.e5

GM Hoyos now faltered against Topalov

with 8...f6?! but he would have had a slightly
preferable position after 8...Nf5.
8...Nf5
(8...f6?! 9.exf6 exf6 10.0-0² Topalov-Leon
Hoyos, Mexico 2010)
9.0-0 Qb6 10.Qa4 e6=

Aside from transposing to the main line of

the Alapin subchapter with 5.exd5, this is White’s
only serious attempt at playing for an advantage.
However, I prefer Black in the tense struggle
which ensues.
(5.dxc5 This offers White no advantage

189
 Black is ready to set up the ‘V’-structure
with ...h7-h5, and it is unclear whether White has
anything other than defensive ambitions.
4...Bg7
5.0-0
5.d4 Qa5! 6.Bxc6 bxc6 7.0-0 Nf6
(7...cxd4 8.cxd4 Nf6 9.Nc3 0-0 10.Bg5 h6
11.Bd2 Qb6 12.b3 d6 13.Rc1²)
8.Re1 0-0
190
9.e5
(9.Bg5 h6 10.Bh4 cxd4 11.cxd4 g5 12.Bg3
Nh5„)
9...Nd5 10.c4 Nc7 11.dxc5 Qxc5 12.b3 d6
13.Bb2 Bg4=;
5.Bxc6?! This is premature, as Black
planned to expend a tempo next move with ...a7-a6
inducing White to take on c6. 5...dxc6=
5...a6!
I very much like this move, which grew up
in the 1960s in the arsenals of grandmasters
Mednis, Taimanov, and Pirc. Black demands a
decision from White, and the ‘wasted’ tempo on
...a7-a6 is not really a waste when we consider that
White rarely couples Bb5xc6 with c2-c3 in the
Rossolimo.
6.Bxc6
6.Ba4 b5 7.Bc2 d5 8.d4 cxd4 9.cxd4 Bg4=
 6...dxc6! 18.Nc4 Nxd4 19.Bxe7 Qd5 20.Bxf8 Rxf8
21.Nxd4
White wishes the c-pawn were on c2; on c3 (21.Ne1 c5³)
it obstructs the natural square for the knight, and 21...Bxd4=;
also weakens the d3-square. The only way for 7.d4?! cxd4 8.cxd4 Bg4 9.Nbd2 Qxd4!
White to make sense of his position is to play for 10.Qb3 Qd7 11.Ng5 Nh6 12.f3 Qd4+ 13.Kh1 Bc8
d2-d4. 14.Rd1 Qa7³
6...bxc6 7.d4 cxd4 8.cxd4 Nf6 9.Nc3 0-0 7...Bg4!? 8.h3 Bxf3 9.Qxf3 Nf6
10.Re1 d6 11.h3 Rb8 12.Qc2 Qb6 13.b3²
7.d3 9...e5 10.a4 a5 11.Na3 Ne7 12.Be3 b6
13.Nc4 0-0 14.Qg3 f6 15.Rfd1²
7.h3 Qd3! 8.Re1 Nf6 9.Re3 Qd8 10.d4 0-0 10.Rd1 0-0 11.d4 cxd4 12.cxd4
11.Nbd2 cxd4 12.cxd4

12...Ne8! A thematic idea for this structure: 12...Ne8!

Black has his eye on the to-be d5 hole after an
eventual ...f7-f5. From e8, the knight can reach d5 A recurring theme.
either via e8-d6-b5-c7-d5 or directly e8-c7-d5. 13.Be3 Nd6 14.Nd2 e6
13.b3 Nc7 14.Nc4 f5 15.Nb6 Rb8 16.exf5
(16.Nxc8 Rxc8=)
16...Bxf5 17.Ba3 Ne6

All of Black’s pawns are on light squares;

as such, he plans to occupy the light squared to-be

191
hole on d5. For his part, White will seek
dark-squared holes like d6 and c5 for his knight,
but Black’s dark squared bishop can protect these
squares, whereas White’s dark-bishop cannot
protect White’s corresponding weaknesses on the
light squares.
14...f5!? 15.e5 Nb5 16.Nb3 e6 17.Nc5 f4
18.Bc1 Nxd4 19.Qe4 Qe7∞
15.Rac1 f5 16.e5 Nb5 17.Qf4

17.Nc4? f4!µ
17...Nc7 18.Nc4 Nd5 19.Qf3 Rf7=

192
 AFTERWORD
Celebrating Raja’s Contributions:
A Year in Review, by GM Romain Edouard
We initiated discussions with Raja about
the idea of a book on the Hyper Accelerated
Dragon during the Reykjavik Open in 2014 (we
noticed his original handling of the opening from
some of his games), and it took us two full years to
finalize a mutually agreeable contract. We had a
great feeling about Raja, who was writing
excellently in our correspondences, well trained by
his studies at Oxford University. Still, he had
never written any chess book and wasn’t a GM yet
(though he has completed his GM norms since
writing his book) — which means he was
relatively unknown to our European audience.
However, after Raja sent me a sample of his work,
I immediately asked our Managing Director,
Daniel Vanheirzeele, to try to find an agreement
with him, as I judged the material outstanding.
I was very happy when a deal was finally
signed, and continued being pleased with the
material I was receiving from Raja. Besides the
great quality of the contents, I was also impressed
how clean his delivered work was — making my
managing editor’s job way easier, as I did not have
to do too many edits before sending the material to
our typesetter! As a company, we are very proud
of the final product: Raja’s book has met and far
exceeded our already high hopes, and reviews
have been overwhelmingly positive.
In addition to being a success for Thinkers,
and a success for Raja, on a more personal note I
have benefitted from Raja’s book in my own
games. I like to be able to change my openings
from time to time, especially from the black side,
but doing so has one big disadvantage: the more
openings you play, the less you tend to remember
about them. Raja’s ‘revisited’ Accelerated Dragon
193
suited me perfectly: not only was it full of new
ideas, and I was pretty sure I would surprise some
of my opponents, but it is also much easier to
remember than most openings.
The reason for this chapter is to show off
the impact that Raja’s book has had in just one
year since its publication. I have selected eleven
games involving Grandmasters, including four of
my own, all played after the publication of Raja’s
book, in variations Raja recommends. Some of
Raja’s most powerful novelties from the book
have now been played, for example his 10...d6! In
the Breyer Maroczy (see Andersen-Demuth 2018)
and 12...h5! followed by 13...e6! in the ‘Panjwani
System’ of the 7.Bc4 variation (see
Petrov-Kapnisis 2017). Still, some of the most
original ideas of the book have not yet been tested
in practice; for example, 1.e4 c5 2.Nf3 g6 3.d4
cxd4 4.Nxd4 Bg7 5.c4 Nc6 6.Nc2 e6!?.
In his Introduction Raja predicted a bright
future for the Accelerated Dragon, and that it
would take its place among other respectable
Sicilians. This seems to be happening already:
who would have thought that Sveshnikov Sicilian
players such as Boris Gelfand would consistently
meet the anti-Sveshnikov 1.e4 c5 2.Nf3 Nc6 3.Nc3
with 3...g6! and enter into a favorable Accelerated
Dragon (since the Maroczy is no longer possible)?
Of course, everything has not been rosy; whenever
a serious theoretical work like Raja’s is completed,
analysts rush to find improvements and new
challenges.
Several new systems for White have been
introduced into practice in many respects as a
response to Raja’s work. For example, one of
these is the so-called ‘10.h3 line’ in the Maroczy:
1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7 5.c4 Nc6
6.Be3 Nf6 7.Nc3 0-0 8.Be2 d6 9.0-0 Bd7 10.h3!?.
Canadian Grandmaster Kevin Spraggett (a former
coach of Raja’s) published some preliminary
analysis on his blog, and readers are invited to
search for improvements for both sides. I have
recently played this line myself with White. When
I asked Raja his thoughts on this system he
admitted it is currently a serious challenge (though
not to the Breyer of course) but he also said that
this is just ‘business as usual’ in research (chess or
otherwise): for a certain period one variation looks
threatening, then computers continue to improve
and new ideas are found to combat it, as well as 12...Bb7
different move order nuances.
Raja also pointed to the recent trend in the I forgot that 12...e6 was Raja’s
Maroczy for Black players to play an early recommendation here, but my move is fine as
...Qd8-b6 as in 1.e4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 well.
Bg7 5.c4 Nc6 6.Be3 Qb6!? with the idea that after 12...e6 13.a3 Bb7„ Raja gave this variation
7.Nb3 Qd8 White will either play 8.Nc3 and allow with the remark that “Superficial developing
8...Bxc3+!? or play 8.Qd2 in which case the 10.h3 moves on White’s part have allowed Black to take
line is no longer playable since the Queen goes to over the initiative: ...d7–d5 is now imminent”.
c2 in that line (so far White players have not tried 13.Qf2 b4 14.Na4 d5!
8.Qc2!?). He understands that many Black players
will be unhappy with a repetition out of the Black is already on the favorable side of
opening, but Raja has little sympathy for them, equality.
and he thinks that if White is strong enough and 15.exd5 Nxb3 16.d6 Qxd6
deeply prepared enough to play the 10.h3 line
successfully (it isn’t a piece of cake to convert the 16...exd6 Computers prefer this with a
slight edge White gets in those lines!) then Black slight edge for Black: 17.Nxb3 Nd5 18.Bd4 Bxd4
should not fear a repetition against such an 19.Rxd4 Rfe8³.
opponent (and such an opponent probably doesn’t 17.Nxb3 Qc7
want a repetition either!).
Take a look at some important games (in
chronological order) that were played throughout
the last year thanks to Raja’s book. You will also
discover a few slight improvements I’ve pointed
out. Enjoy!
1
K. Drozdowski (2461)
R. Edouard (2622)
Warsaw 2017

1.e4 c5 2.Nf3 Nc6 3.Nc3 g6 4.d4 cxd4

5.Nxd4 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.f3
Qc7 10.Qd2 b5 11.0-0 Na5 12.Rad1
18.Nac5?!

¹18.c3 bxc3 19.Nxc3 e6!=

18...Bc6

18...Nd5µ
19.Nd3 Nd5 20.Bd4 e5 21.Bc5 Rfb8³

White’s position is very loose.

22.Nd2?! Bb5 23.Ne4 Rc8 24.Rfe1 f5!
25.Bd6 Qxc2 26.Rd2? Qc6

194
 14.Qf3 Nf4

14...a4 15.Ba2 Qe5³

15.Rad1

White is losing a piece.

27.Nxe5 Bxe5 28.Bxe5 fxe4–+ 29.fxe4
Ne7 30.Rd6 Rf8 31.Qe3 Qc4 32.h4 Nc6 33.Bf6
Qf7 34.h5 Qa7 35.hxg6 hxg6 36.Qxa7 Rxa7
37.Rc1 Nb8 38.e5 Nd7 39.Be7 Re8 40.Rxg6+
Kf7 41.Bc5 Rc7 42.Rg5 Rxc5 43.e6+ Rxe6 15...Ne6?!
44.Rgxc5 Nxc5 45.Rxc5 Re1+ 46.Kh2 Re2
15...a4 16.Ba2 Be6³
0–1 16.Bxe6! Bxe6 17.Qe3

2 A few moves ago White’s pieces were

D. Vocaturo (2592) awkwardly placed but now White has nice
M. Colpe (2412) harmony.
Helsingor 2017 17...Rfb8 18.b3! a4 19.Ne2 axb3 20.cxb3
Bb2?! 21.Rd2 Bxa3 22.b4±

1.e4 c5 2.Nf3 Nc6 3.Nc3 g6 4.d4 cxd4

5.Nxd4 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.0-0
Qc7 10.h3 b5

22...Bc4 23.Rfd1 Rd8 24.Bb6?

24.Nd4!± e5? 25.Nf5! Rxd2 26.Qxd2

gxf5? 27.Qg5+ Kh8 28.Be7+–
11.Nxc6?! dxc6 12.Bc5 a5 13.a3 Nh5! 24...Rxd2 25.Rxd2 Qe5 26.Bd4 Qc7
195
27.Bb6 Qe5 28.Bd4

14.gxh5!?

Black had every reason to avoid repetition White is the first to deviate from Raja’s
and play on himself. After all, Black is up a pawn. analysis in the book.
28...Qc7 14...Nxh5 15.Nde2 Ne5³ 16.Rh3?!

28...Qb8! 29.Qc3 f6! 30.Ng3 Bf7 White

can probably hold but Black is certainly to be
preferred here.
29.Qc3?!

29.Bb6=
29...c5?

29...f6! 30.Ng3 Bf7³

30.Bxc5± Qb7 31.Ng3 f6 32.Rd1 Ra4?
33.e5!+– Kf7 34.exf6 exf6 35.Re1 Be6 36.Ne4

1–0

3 16...Rfc8
M. Petrov (2426)
S. Kapnisis (2469) 16...d5! 17.exd5 exd5µ 18.Nxd5? Bxd5
Athens 2017 19.Bxd5 Rfd8! 20.Qb4 Rac8 21.Nc3 Nf6 22.Qb3
(22.Bb3 Rxd1+ 23.Kxd1 Qd7+–+) 22...b4–+
17.Bd4 Qb8?!
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.f3 Too slow. Very strong was 17...b4 18.Na4
Qc7 10.Qd2 b5 11.0-0-0 Bb7 12.h4 h5 13.g4 e6 a5! 19.Nb6

196
 1.c4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Nc6
5.e4 Bg7 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Bd3 0-0 11.0-0 b6!?

19...Ba6!µ 20.Nxa8?! Rxa8 21.Kb1 Bxe2

22.Qxe2 Nf4–+.
18.f4?!

18.Qg5!?²
18...Nc4 As I had tried the more solid 11...d6
earlier, I decided to surprise my opponent with this
fresh alternative suggested by Raja, which I
considered very reliable.
12.Ne2 Ne6 13.Qd2 Bb7 14.f3 f5 15.Rad1
Qf6

19.Bxc4 Rxc4?

19...Bxd4! 20.Nxd4 Rxc4=

20.Bxg7 Kxg7 21.b3 Rc8 22.Qd4+ Kg8
23.f5± Qa7 24.Qd2 Kg7 25.fxg6 b4 26.gxf7 bxc3
27.Qg5+ Kf8 28.Rxd7 The tension has been building, things are
about to explode.
1–0 16.c5?! f4

4 a) 16...Nxc5 17.Bc4+ Kh8 18.b4 Ne6

W. Spoelman (2586) 19.Qxd7±;
R. Edouard (2607) b) 16...bxc5 17.Bc4 d5∞ (17...Rfd8
Belgium 2017 18.Bg5+–)
17.Bc4?
197
 13...a6!

13...Qb6?! 14.Qd2 Rfc8 15.Rfc1 a6 16.a5

Qc5 17.b4 Qc6 18.Nxd4 exd4 19.Bh6± Horvath,A
(2473)-Seres,L (2461) Heviz 2011.
14.Nxd4 exd4 15.Bd2 Rc8 16.Rc1

17.Bf2 bxc5 18.Bc4 Rfd8 19.Nc3°

17...d5!! 18.c6 dxc4 19.cxb7 Rad8µ
20.Qc2?!

20.b8=Q! Rxd2 21.Qxf8+ Bxf8 22.Bxd2µ

20...fxe3 21.Qxc4 Qf7 22.Rxd8 Nxd8
16...f5
0–1
16...Qd7 17.a5 f5 18.exf5 Bxf5 19.Bb4 h6
5 20.Qd2 Kh7 21.Rce1 Rfe8 22.b3 d5!= Arsovic, Z
J. Ferreira (2499) (2446)-Popovic,M (2401) Vrnjacka Banja 2010.
R. Edouard (2607) 17.exf5 Bxf5 18.Qb3 Qd7 19.Bxf5 Qxf5?!
Belgium 2018
19...Rxf5 20.c5+ d5

1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6

5.c4 Bg7 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Bd3 0-0 11.0-0 d6 12.a4!? Be6
13.Nb5

21.Rfe1 d3!=
20.c5+ Rf7 21.cxd6 Rxc1 22.Bxc1 Be5²

½–½

6
198
 O. Kurmann (2467) Bf4 is why Black must sacrifice with 15...Nxe4!!]
A. Demuth (2547) 15...Nxd5 16.Qxd5 Bxb2 17.Rd1
Bad Ragaz 2018 (17.Rb1?! Ba3 18.0-0! Be6µ) 17...Be6 18.Qxc5
Qxc5 19.Bxc5 Bf6 20.Bxe6 fxe6=
12...Nxd5 13.exd5 Na5µ 14.Qxb4 Nxb3
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6 15.Qxb3 Rb8 16.Qa2
5.Nc3 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.f3
Qc7 10.Qd2 b5 11.a4 b4

16...Qa5+

12.Nd5?! 16...e5! 17.dxe6 dxe6 18.c3 e5 19.Ne2 Be6

20.Qb1 Qc4µ
Recall Raja’s analysis after 12.Na2 Rb8! 17.Ke2 Qb4
13.Nxc6
[13.0-0 Nxd4 14.Bxd4 Ne8! 15.Bxg7 17...Bb7 18.c4 e6!µ
(15.Nxb4 Bxd4+ 16.Qxd4 e5 17.Qc3 Qb6+–+) 18.Rhd1
15...Nxg7= The knight will find its way to c5 via
e6; White’s pieces are sloppy on b3 and a2.]
13...dxc6 14.Nxb4 c5 15.Nd5
[15.Nd3

18...Qxb2?!

18...Qd6! 19.h3 e5 20.dxe6 fxe6 21.Kf2

15...Nxe4!! (15...c4 16.Bf4±) 16.fxe4 c4 Qh2 22.Rh1 Qc7µ
17.Bf4 e5³ The ability to play ...e5 in response to 19.Qxb2 Rxb2³ 20.Kd3 a5 21.Rdb1 Ba6+
199
22.Nb5 Rb8 23.Rxb2 Bxb2 24.Rb1 Bxb5+
25.axb5 Rxb5 26.Bd4 Rxd5 27.Rxb2 e5 28.c3 14...Rc8 Raja gives this as his main line
exd4 29.cxd4 but also offers 14...Qd7 as a safe alternative.
15.Kh1
And Black went on to win the endgame.
29...Kf8 30.g4 Ke7 31.Rb8 a4 32.Ra8 This move was not analyzed by Raja, he
Rb5 33.Rxa4 Rb2 34.h4 Rh2 35.Ra6 Rxh4 considered 15.Nd5, 15.f4, and the novelty 15.Ne2
36.Ke3 h5 37.gxh5 Rxh5 38.Ke4 Rh4+ 39.Ke3 which he gives an ‘!’.
Rh1 40.Ke4 Re1+ 41.Kf4 d6 42.Ra7+ Kf6 15...b5!? 16.Nd5 Bxd5 17.cxd5 f5
43.Ra5 Ke6 44.Rb5 f5 45.Kg5 Rg1+ 46.Kf4 Rg2
47.Ra5 Rd2 48.Ke3 Rb2 49.f4 Rb3+ 50.Ke2 Rb6
51.Ke3 Rb7 52.Kd3 Rb4 53.Ke3 Rb6 54.Kd3
Kf6 55.Ra8 Rb3+ 56.Ke2 Rb7 57.Kf3 Rb1
58.Ke2 Ke6 59.Ra5 Rb8 60.Kd3 g5 61.fxg5 Rg8
62.Ke3 Rxg5 63.Ra4 Rg3+ 64.Kf4 Rg4+ 65.Ke3
Re4+ 66.Kd3 Re1 67.Ra5 Rh1 68.Ke3 Re1+
69.Kd3 f4 70.Kd2 Re4 71.Kd3 d5 72.Ra8 Re3+
73.Kd2 Kf5 74.Rf8+ Ke4 75.Kc2 Ra3 76.Kd2 f3
77.Re8+ Kxd4 78.Rd8 Ra1 79.Rf8 f2 80.Rf4+
Kc5 81.Rxf2 Ra2+

0–1

7
L. Bruzon Batista (2673) 18.Bxd4
R. Edouard (2612)
Saint Louis 2018 18.exf5 Nxf5=
18...fxe4
1.Nf3 c5 2.e4 g6 3.d4 cxd4 4.Nxd4 Nc6 An important inbetween move.
5.c4 Bg7 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4 18...exd4 19.exf5 gxf5 Black ’s kingside
9.Qd1 e5 10.Bd3 0-0 11.0-0 d6 12.Qd2 Be6 13.b3 (pawns) are broken. 20.Rde1±
a6 14.Rad1 19.Bxe5 Bxe5 20.Bxe4 Qg4! 21.Rde1?!

21.Qd3 Rac8 22.g3 Rc3 23.Qb1 Rfc8°

21...Qh4=

14...Qd7
200
 22.g3
22.h3 Rf4 Black has full compensation: in
such opposite color bishop positions, the pawn
count is less important than the activity of the
pieces. 23.Bc2 Raf8 24.Re2 Bd4 25.Kg1 Bc5=
22...Bxg3! 23.fxg3 Rxf1+ 24.Rxf1 Qxe4+
25.Qg2
25...Qe7 26.Qd2 Qe4+ 27.Qg2 Qe7
28.Qd2 Re8
I decided to play on... it is drawish of
course but Black is more comfortable.
29.Kg1 Qe5 30.Rc1 a5 31.Qd3 b4³
201
32.Rc2 Qe3+?!
I could have instead posed serious
problems with 32...Rf8! 33.Rd2 (33.Rf2 Qa1+
34.Kg2 Rxf2+ 35.Kxf2 Qxa2+µ) 33...Rf5! 34.Kg2
h5 35.Qd4 Kg7 36.Qxe5+ Rxe5µ White’s rook is
passive; Black will probably end up winning the
d5 pawn and retain practical winning chances.
33.Qxe3 Rxe3 34.Kf2
34.Rc8+ Kg7 35.Ra8=
34...Rd3 35.Ke2 Rxd5 36.Rc7
Since White has managed to cut off Black
’s king, this is a drawn endgame.
36...Re5+ 37.Kf3 h5 38.Rd7 Rd5 39.Ra7
g5 40.g4 hxg4+ 41.Kxg4 Kf8 42.Kf3 Ke8 43.Ke4
Re5+ 44.Kd4 Kd8 45.h3 Kc8 46.Kd3 Kb8
47.Rd7 Rd5+ 48.Kc4 Kc8 49.Ra7 Rc5+ 50.Kd4
Re5
½–½
8
Y. Zherebukh (2633)
R. Edouard (2612)
Saint Louis 2018
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6
5.c4 Bg7 6.Be3 Nf6 7.Nc3 d6 8.Be2 0-0 9.0-0
Bd7 10.Nb3 a5 11.Nd2 Bc8 12.Rc1 Nd7 13.Nb3
a4 14.Nd4 Nxd4 15.Bxd4
 18...Bxf4! This is the best way for Black to
equalize. In general Black is happy to exchange
these dark-squared bishops; as long as White has
nothing immediate then Black’s sounder pawn
structure and dark square control will tell.
[18...Qb6+?! This was Raja’s
recommendation, which has now been corrected.
19.Kh1 Bxf4 20.Rxf4 Qxb2 21.Rc2 Qb6

So far we have been following Raja’s

analysis. Here, over the board, I decided to deviate
from what I knew was Raja’s recommendation,
15...Bh6, because it seemed to me that his analysis
led to a suspicious position. It happens that I was
right — White has a crushing resource which Raja
missed, but it turns out I should have played
15...Bh6 anyway and deviated from Raja’s
analysis a bit later. 22.Nd5! This move, with the idea of
15...Bxd4 23.c5!! was missed by Raja. (22.Rd2 Ne5 23.Nd5
Qc5 24.Nf6+ Kg7 25.Rxd6 Be6= Panjwani)
I decided to defend this slightly worse 22...Qd8 23.c5!! dxc5 (23...Nxc5 24.Nf6+ Kg7
endgame instead of risking the complications after 25.Qa1! Ne6 26.Rf3±) 24.Bb5 Ne5 25.Rxc5 Kg7
15...Bh6. This turned out to be an easier task than 26.Qa1 Qd6 27.Rc7 Be6 28.Rf6 Kg8 29.h3!!±
it should have been, with accurate play White can
pose serious problems here. However, as Raja
insisted throughout this book, computers make
everything look easy, and in practice even strong
grandmasters do not play accurately enough to
pose problems with these slight advantages.
15...Bh6! 16.f4 e5 17.Be3 exf4 18.Bxf4

Apparently the only move which gives

White an advantage, but it is a clear advantage.
Although this is a computer move, Black has been
struggling with an unpleasant position for several
moves now.]
19.Rxf4 Ne5!

202
 20.Nd5 (20.Rf1 Be6 21.Nxa4 Qa5 22.b3
Nxc4 23.Bxc4 Bxc4 24.Rxc4 b5 25.Rd4 bxa4
26.bxa4 Rab8=; 20.Nxa4 Bd7 21.Nc3 Qb6+
22.Kh1 Qxb2 23.Rc2 Qa3³) 20...Be6 21.Nf6+ Kg7
22.Qd4 Qc7=
16.Qxd4 Qb6
22...f6 23.exf6 exf6 24.Qh4 Kg7 25.Nxc5
h5 26.h3 Bf5 27.Nxa4 Qd4 28.Rc7+ Kh6 29.Qxd4
Rxd4 30.Nc3²
17...Nxb6 18.f4
18.Nd5 Nxd5 19.cxd5 e6 20.Rc7 exd5
21.exd5 Bf5! 22.Rxb7?! Rfb8 23.Rb5 Bd7!
24.Rxb8+ Rxb8³
18...Be6 19.Nb5 Rfc8 20.b3 axb3 21.axb3

17.Qxb6

17.Qd2! Nf6 (17...Nc5 18.Nd5 Qd8

19.Rfd1 Be6 20.Qd4 Bxd5 21.cxd5²) 18.c5!! dxc5
19.e5 Rd8 20.Qh6 (20.Qg5 Ne8 21.Ne4 Be6
22.Nxc5 Ng7 23.Qxe7 Qxb2 24.Nxe6 Nxe6=)
20...Ng4 21.Bxg4 Bxg4 22.Ne4
21...Bd7! 22.Nd4 Ra2 23.e5 dxe5 24.fxe5
e6! 25.Bf3 Bc6= 26.Nxc6 bxc6 27.Rfd1 Ra7
28.Rd6 Rac7 29.b4 Kf8 30.c5 Nd5 31.Bxd5
½–½

9
Wei Yi (2734)
S. Shankland (2671)
Liaocheng 2018

203
 1.Nf3 c5 2.e4 Nc6 3.Nc3 g6 4.d4 cxd4
5.Nxd4 Bg7 6.Be3 Nf6 7.Bc4 0-0 8.Bb3 a6 9.0-0

Targeting the d5 pawn but also threatening

a minority attack with ...b5–b4.
19.Nc2 Nc4 20.Bc1 a5 21.Qg3 Rc5
22.Ne3 Nxe3 23.Bxe3 Rc4µ
9...d6
23...Rxd5?? 24.Qf3+–
Raja recommends the move order 9...Qc7 24.Bg5 Bxg5 25.Qxg5 b4 26.cxb4 Qxb4
to keep the option of keeping the pawn on d7 and
opting instead for a Taimanov structure with 26...Rxb4 27.b3 a4 28.bxa4 Rxa4 29.Rd2
...e7–e6. Ra5µ
9...Qc7 10.f4 d6 11.h3 Na5 12.Qd3 b5³ 27.Qd2
This line given by Raja is similar to the game
though White had to play f2–f4 to induce ...d7–d6
by Black, whereas Shankland played it voluntarily.
10.h3

10.Qd2 Ng4 11.Nxc6 bxc6 12.Bg5 a5

Black should be okay.
10...Na5!

Black is at least equal, but his position is

much easier to play so practically speaking Black
is bound to get a slight edge in the coming moves.
11.Qd3 b5 12.Nd5 Bb7 13.c3 Rc8
14.Nxf6+ Bxf6 15.Bh6 Re8 16.Bd5 Qd7 17.Rad1
Bxd5 18.exd5 Qb7³
27...Qxd2?!

Unnecessarily direct.
27...Rec8 28.Qxb4 axb4 29.Rd2 Rc2
30.Rfd1 Kg7µ
28.Rxd2 a4 29.Re1 Kf8³

Black only lost this game because he took

serious risk in his attempt to win.
204
 30.Re3 Rb8 31.b3?! Rcb4 32.Rdd3 R8b5 of days before, and we had a chat before that
33.g3 h5 34.h4 Rc5 35.Rd2 axb3 36.axb3 game!
10.Nb5
Although the computer assesses the
position as µ, it is not easy to see how Black
makes progress.

10...d6!N

Shank land slowly loses his way, I give the This is the first published game with Raja’s
rest of the game without notes since this is not novelty. As he mentions in his book though, he
meant to be a discussion on rook endings. played it twice in unpublished tournament games
36...Rb7 37.Kg2 Rcb5 38.Rdd3 Ke8 (wins) against GM Gorovets and WGM Nemcova.
39.Rf3 Rc5 40.Rd2 Rbb5 41.Rfd3 Kd7 42.Rf3 f6 11.Nxd4 exd4 12.Bxd4 Qa5+ 13.Ke2 Be5
43.Rfd3 Ke8 44.Kh2 Kf7 45.Kg2 f5 46.f4 Kf6 14.f3 f5
47.Kf2 Rb4 48.Re2 Re4?! 49.Kf3 Rxe2
50.Kxe2= Rc2+ 51.Ke3 e5 52.dxe6 Kxe6 53.Kd4
d5? 54.Rc3!± Rd2+ 55.Ke3? Rg2 56.Kd4 Rd2+
57.Kc5! d4 58.Rf3 Rd1 59.Kc4 Kd6 60.Rd3 Re1
61.Rxd4+ Kc6 62.Rd3 Re4+ 63.Rd4 Re3 64.Rd3
Re4+ 65.Kc3 Kc5 66.Kd2 Kb4 67.Re3 Rd4+
68.Rd3 Re4 69.Rd6 g5 70.hxg5 h4 71.gxh4 Rxf4
72.h5 Rh4 73.h6 Rh5 74.Rd8

1–0

10
M. Andersen (2589)
A. Demuth (2538)
Berlin 2018
15.exf5
1.c4 c5 2.Nf3 g6 3.d4 cxd4 4.Nxd4 Bg7 15.Qd3 Raja gives this as his main line.
5.e4 Nc6 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4 15...Bxf5 16.Kf2 0-0-0
9.Qd1 e5
16...0-0!? Panjwani.
I can testify that Adrien played this line 17.Rc1
thanks to Raja. He had bought the book a couple
205
 17.Be2? Bc2!³ Panjwani.
17...Kb8 18.Be2 Rhe8

22...Rde8?!

a) 22...Be6 23.Qf4 g5 24.Qg3± (if Black

Until this point both players have been had a pawn on h5, as in the variation 19...h5, then
following Raja’s analysis. Here Andersen deviates. there would now be ...h5–h4);
19.Re1!? b) ¹22...Ree8!=
23.b4
19.Rc3 d5„ Panjwani
19...d5?! 23.Qd4!²
23...Qb6+ 24.Kf1 Bd7 25.Qf4 g5 26.Qd2
a) 19...h5!? 20.Rc3 (20.h3 g5!„) 20...d5
Now if the variation proceeds as in the main game
then... 21.Bxe5+ Rxe5 22.Qd4 dxc4 23.Qxc4 Be6
24.Qf4 g5= 25.Qg3?? h4–+;
b) 19...Bxh2 Black can safely regain his
sacrificed pawn. 20.Rc3 (20.g3 Bc8! 21.Kg2
Bxg3! 22.Kxg3 Qg5+ 23.Kf2 Qh4+ 24.Kg1 Qg3+
25.Kh1 Re6 26.Bf1 Qh4+ 27.Kg1 Rxe1 28.Bxa7+
Kxa7 29.Qxe1 Qg5+³) 20...Be5 21.g4 Bc8 22.Ra3
Bxd4+ 23.Qxd4 Qc5=
20.Bxe5+ Rxe5 21.Qd4 dxc4 22.Qxc4

26...Rxe2! 27.Rxe2 Bb5 28.Rce1 h6=

29.a4 Bc4 30.a5 Qb5 31.Kf2 Bxe2

½–½

11
A. Pichot (2564)
R. Panjwani (2456)
Montevideo 2018

206
 19.Bxe5 dxe5
For the final game of this chapter I decided
to include one of Raja’s from the recent
tournament where he achieved his final
Grandmaster norm. Even after publishing all his
analysis for the public, he still trusted his pet
system in an important encounter with GM Pichot!
1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 g6
5.c4 Bg7 6.Be3 Nf6 7.Nc3 Ng4 8.Qxg4 Nxd4
9.Qd1 e5 10.Bd3 d6 11.0-0 0-0 12.Qd2 Be6

20.a4 (20.Qxb5 Qd6°; 20.Bxb5 Bd7 21.a4

Rab8³) 20...Rfc8! Black plays in Benko-gambit
fashion. 21.axb5 (21.Bxb5 Rc3 22.Rab1 Qc7³)
21...Kg7 22.Ra4 Rab8 23.Bc4 Ra8= 24.Rfa1?
Rxa4 25.Rxa4 d3!µ
16...d5! 17.Bxd4 exd4 18.Nf4

18.Nxd4 Bxd4 19.Qxd4 dxe4 20.Qxd8

Rfxd8 21.Bxb5 (21.Bxe4 Ra3!=) 21...Rd2 22.Ba4
Rc8=
13.b3 18...dxe4 19.Nxe6 fxe6 20.Bxe4

This is the first deviation from Raja’s

analysis. White does without the usual rook moves
(13.Rad1, 13.Rac1, etc.).
13...a6 14.Ne2 b5! 15.cxb5 axb5

20...Qa5!=

20...d3∞
21.Qe7 Rfe8 22.Qg5 b4! 23.Qg4 Rac8
24.Bd3 Rc3 25.Bc4 Qf5
16.Qb4
25...Kh8! Black can try for an advantage:
16.Nxd4 exd4 17.Bf4 Qb6 18.Qb4 Be5 in such opposite colored bishop endgames it is
207
often favorable to sacrifice a pawn for activity. ½–½
26.Bxe6 d3 27.Rad1 Rd8 28.Bc4 d2³
26.Qxf5

208
209