Sunteți pe pagina 1din 23
THE MATHEMATICS OF GAMBLING By Dr. Edward O. Thorp Table of Contents SECTION ONE—CARD GAMES 1 Chapter 1: Introductory Statement. . Chapter 2: Blackjack. Chapter 3: Baccarat SECTION TWO—THE WHEELS 4 Chapter 4: Roulete....... Chapter 5: The Wheel of Fortune SECTION THREE—OTHER GAMES. Chapter 6: Horse Racing . Chapter 7; Backgammon « SECTION FOUR—MONEY MANAGEMENT ........ ‘Chapter 8: Mathematical Systems ..........-.1I3 Chapter 9: Optimal Betting .... Appendices «... Keeping Your Gaming Knowledge Current ......14L About the Author ‘Edward Thorp i adjunct professor of finance and mathematios atthe University of California at Irvine, where he has taught courses in finance, probability and functional analysis. He previously taught atthe University of California at Ls Angeles, the Masachusens Instnte of Technology and New Mexico Ste University ‘Thorp’ interest in gambling dates back almost 30 years, while Ihe was tll in graduate school at ICT-A. Tt was bere that he fist formulated his dream of making money from the development ‘ofa scientifically based winning gambling systema, His first sub- Jest of study was the roulewe wheel, which offered him the op- Foray fuse modern pss o pect te resin place With the roulete work unfinished, Thorpsatenton was diverted by the blackjack work of Raldwin, Cantey, Mrizel and MeDer- ‘ott. He st to work on this new problem. With the ad of com- puter, Thorp developed the basic strategy and the five-couat, en- ‘count and ultimate counting strategies. He used these methods with sucess inthe Nevada casinos. The work was first publiciz- ed in a scientific journal and saw broad public exposure inthe 1962 book Beat the Dealer. The book underwent a revision in 1966 aid it ill epated as the lassi aly wrk a Ue “ack Jack revolution” which continues to this dey nthe late 1960s, Thorp developed with Sheen Kassout a sue- cessful method for stock market investing involving warrants that proved 50 profitable that Thorp turned $40.00 into $100.00 in two years. The strategy was publised in Beat the Market in 1967 ‘Additionally, using this strategy and further eefinements, Thorp ‘naiages ul-nilionllasiavesineat portfolio. He is Pres ent of Oakley Sutton Management Corp, and Chairman of the the Board of Oakley Sutton Securities Corp. “Thorp has continued to advance new theories for eambling and other games, as wel asthe stock market. Section One Card Games Casino card games such as baccarat and blackjack ) ¢-z 914%, £2 10%, The Mathenaics of Gambig is win rates en units pe hour and his verge bet SURG a’ Civen thse assumptions, being chested ten Heer hour ov onetenh ofthe time would ene hi advan- {ape Beng cheated more than ten pact ofthe time would prob- Sj fimino ate regi the real words probably moreefTectvethanintne typothetial example just ted, because te calculations fr that ‘Rane assure cheating equal kl fr small bes and big Seas my experience, the Bettor Se riuch more Healy to be hesied on lage bes than on smal ones. Therefor, the dar tho cheats wih mimi effeney wil wait uni a player Thakes his top be, Suppose that bet total five nis. Ifthe cheat Shifts the odds to 80 perent in favor Of the house, ie expected Tosis 2-1/2, ands four cheaing efforts per 100hands wil ‘acl a profesional players advantage. A cheating rate of ive ‘rte hand per 100 pt this player aa eevee dcadeansage ‘We can sce fom his that comparavely small amount of cheating applied to the larger hands can have a significant impact onthe game’s outcome. This gives you an idea of what tolook for Shen you ar in he casinos and nk tat someting may be Missing Cards: The Short Shoe have heard complaints that cards have been missing from the ‘pack in some casino blackjack games. We'll discuss how you ‘ight spot this cheating method. In 1942, | wrote on page SI of Beat the Dealer, “Counting the. card, is an invaluable asst in the detection of cheating ‘because a comnnion devices to remove one or more cards from the deck." Lance Humble discusses cheating methods for fourdeck games dealt from a shoe in his Intemational Blackack Cub newsletter. He say, “The house can take certain cards such a8 tens and aces out of the shoe. This is usually done after several rounds have been dealt and after the decks have been shuffled several imes. Its done by paiming the cards while they are being 2 Blokick shuffled and by hiding them on the dealer's person, The dealer lien disposes Of the cards when he goes oh fis break.” But cheating this way is not limited tothe casino, Payers have been known to remove “small” cards from the pack to tit the edge their way, The casino can spot tis simply by taking the pack and ‘counting it; the player usually has to use statistical methods. In the cheating trade, the method i known asthe short shoe. Lets say the dealer is dealin froma shoe containing four decks ‘of 92 varus each In 32 cards, there should be 16 ten-value cards: ‘the ens, jacks, queens and kings. Logically, in four decks of 208 cards, there should be 64 ten-value cards” I'l call all ofthese ens" from now on. Casinos rarély remove the aces—even ‘novice players sometimes count these. ‘Suppose the sift boss or pit boss takes outa total of ten tens, some ofeach kn, of course, not all kings o queens The shoes shottened fu 6 ves wo 34 tens, and the four decks from 208 cards to 198 card ‘Thelossof these ten tens shits the advantage from theplayer to the dealer or house. The ratio of exhers/tene changes from the sonal 14/64 =2.3510 144/54 =2.67, and this gainsalitle over ‘one percent forthe house. How can you discover the ack of tens | without te dealer knowing it? __ Hewes one method tha is sed. If you're playing a the black- jack table, stn the last chairon the dealer's right. Bet asmal fixed ‘amount throughout a whole pack of four decks. After the deer ts the ct card back ony, ler's say, ten percent ofthe way into the four shuffled decks and returns the decks into the shoe then | ready yourself to count the cards. Play your hand mechanically, only pretending interest in your good or bad fortunes. What you'einerested in finding outs the numberof tensin the whole Four-deck shoe, ‘Les say the shift boss has removed ten tens. Reports are that they soem to love removing exactly ten from a four-deck shoe) ‘When the white eut card shows atthe face of te shoe, let's say that the running count of tens has reached $2. That means ‘mathematically that if all 64 ten-value cards were in the shoe, a The Mathematics of Gambling then, ofthe remaining 15 cards behind the cut card, as many as 12 of them would be teas, which mathematically is very unlikely. This is how one detects the missing ten tens because the dealer never shows their faces but just places them face dawn on top of thestack of ducarded cardstohisright, which then proceedsto shuffle face down in the usual manner preparatory to another ourdeck shoe session. Although at first the running counts not easy to keepin areal casino situation, a secondary ciTiculy is estimating the appror- mate number of cards let behind the cut card aftr al the shoe bas been dealt. To practice this, take any deck of S2cards and cut off what you think are ten, 18 or 20 cards, commit yourself to some definite number, and then count the cards to confirm the closeness of your estimate. After awhile, you can look atabunch ‘of cards cutoff and come quite lose to their actual number. Insuramary, count theniumber of tens seen trom the Desig of a freshly shuffled and allegedly complete shoe, When the last card is seen and itis time to reshuffle the shoe, subtract the ‘umber often sen from the number that are supposed to be in the shoe—68 fora four-deck shoe—to get the numberof unseen tenwalue cards which should remain. If $8 ten-value cards were seen, there should be ten tens among the unused cards. Then estimate the umber of unseen cards. You have to besuretoadd to the estimated residual stack any cards which you didnot see uring the course of play, such as burned cards. Step fouristoask ‘whether the numberof unseen ten-value cards is remarkably large for the numberof residual cards. If 50, consider seriously the possibility thatthe shoe may be short. For instance, suppose there are 1S unseen cards, ten of which are supposed tobe ten-values. A ‘computation shows thatthe probably that the last 1S cards of a ‘wellshuffed four-deck shoe will havea least ten en-au cards {5 0.003247 or about one chance in 308. Tihs the evidence against the casino on the basis of this one shoe alone isnot overwhelming, Buif we wereto count down the same shoe several times and each ime were to find the remaining cards suspiciously ten-tch, then the evidence would become very Pa lacjack ssrong. Suppose that we counted down the shoe four times and ‘har each time there were exactly 15 unseen card. Suppose that the numberof unseen tens, assuming a ful our decks, was nine, 1, ten, and [3 respectively. Then referring to Table 2-4, the pro- bbls to six decimal places are H() = 14651 wo have nine ‘or more unseen tens, and fr at last I, ten, and 13 respectively, the chances are Hl) = .000539, HO) = .003247, and H(3) "= 000005, These correspond io odds of about 168, 1855, V308 and 17200000 respectively. The odds against al ese events hap- ‘ning ogether is much greater sil. In ths eaample, the evidence ‘Sonal suggests that upto nine ea-alue cards are missing, There can't be more than nine missing, of course, because we sw all ‘but nine on one countdown. ‘If the casino shuffles after only 104 cards are seen, itis not so easy totellif ten ten-value cards were removed. A mathematical ‘roof of this fs contained in the appendix.» “This discussion should makeit clear thatthe method suggested is generally not able to easy spot the removal often-value cards| ules he shoes counted sea times os del down clos to ‘One of the interesting ironies of the short shoe method of cheating playersis that neither the shift boss nor the ptboss—the laver bringing the decks of cars the dears ble—ueed el the dealer that his shoe s shor. Thus, the dealer doesn’t necessarily hhavetoknow that he'scheatng. Afterall he's jut dealing. I'san {pe question how many deals know ta the ding from ‘Reports are that the short shoe is a frequent method that, casinos use in cheating at blackjack using more than one deck. “The tables with higher minimums (say $23) are more tempting ‘candidates for short shoes than those with the lower minimums. “An experienced card counter can improve the method by count- ing hoth tens and non-iens. Then hell now exactly how many unseen cards there are, aswell as unseen tens. Table 2-4.can then ‘be use with greater confidence. Inpractce, you don’t need to count through a shoe while bet- 2 see Arent pe The Mathematics of Ganbing Table 24 ting (and thus losing money in the proces) to find out that the ‘casino cheating. Ifyou suspect foul play, count whe sanding ‘Behind the player tothe dealer's right "You might easy catch a short shoe by simply counting all the cards that are used, whether or not yousee what they are, Then if the remaining cards, atthe reshuffle, are few enough so you can 6 Buactjack ‘accurately estimate their number, you can check the total count. For istance, you count 165 eards used and you estimate dat 31 ‘& 3cardsremain. Thenthere were 196 = 3 cards athe than the 208 expected, 50 the shoes short. ‘Acting conntermeasireisto put hack a, $or6 fo each ace or ten-value card removed. Then the total number of cards remains 208, and thecasno gets an even greater advantage thanit ‘would froma short shoe. ‘Cards do get added to the deck, and there's u spooky ci «idence oillustrate this. On page SI of Bea the Dealer, rotein 1942, “One might wonder a this point whether casinoshave also rind adding cards tothe deck Uhave only sen it done once. Iis very risky. Imagine the shock and fury ofa player who picks up his hand and ses that not only are both his cards S, but they are also both spades." And then 15 years late in 1977, a player ina ‘one-deck game did peta hand with two ofthe sare Garde Sot spades. Walter Tyminsk's casino gaming newsleter, Rogue et ‘Noir News, reported on page3of the June 15,1977 issue, “What ‘would you dof the player a you right ina single blackjack game had two 5 of spades? Nicholas Zaika, a bail bondsman from ‘Deri ad that experience at the Saharain Las Vegason May 24 ata $5 minimum table. “zatea wasnt inthe best of humor because he had reportedly lost $394,000 at other Sahara tables, by far the largest loss he has ever experienced. Zaika had the blackjack supervisor check the cards and there were 53 cards in the deck, the duplicate be- inthe 5 of spades. The gumer has engaged the services of Las Vegas attorey George Grazadei to pursue claims he feels that be has against the casino. The Sahara denies ayy wrongdoing and says that it i ‘cooperating fully with th investigation... Players arentlikelyt0| {introduce a exzaS because the presence ofthe extra’ favors the Inoue and not the player” ‘Suppose instead of ust counting tens used and totalcards sed, you kept track of how many aces, 25,36, queens, kings, andso on ‘Were used. Ths extra information should give the player a better z The Mathematics of Gambling ‘chance of detecting the short shoe. The ultimate proof would be to count the number of each of the 52 types of cards which have ben used. Mathematical readers might wish to investigate effec- ‘tive statistical or other ways of using information for detecting thoes in which the numbers of some of the cards have been 4 2s Baccarat “The games of baccarat and chemin de fe are wel known gam- ‘bing games playee for high stakes in several pars ofthe world. is sid tobe a card game of Falan org that was troduced into France about 1190 A.D. Two forms ofthe game Aevelped One form wae caled Rocce andthe oer ws alls chemin ofr. The mos basi diference between these tw is simply that chee hands ate dealt n bacarat (called bocarat en bang in England) and two had ae dealt in chemin de fet (Galles baccaratcheman de Yr in England and Nevada) The cards ae trough ine are each wort terface valve and the cards ten jack, queen and king are each worth 20 pins. ‘Aandi evan asthe umn modo tn of i card, 4.8, om. iy the las digit ofthe (oui counted. The objet ofthe game is tobe a lose to eght or ine as posible wth two cards, of 4 clos to nine as possible with at most thee cards if one does othave eight or nine on is st wo cards Then Ge high hand “The games ofbacarat and chemin defer Became popula in » ‘The Mathematics of Gambling ic casios allover Europ, as wells in priate games, about EBD Arthe presen me, one orth of ese games ae well oa in Landon, southern France, the Riviera, Germany and the nied Sten A form of chemin def, which we shal ell ewe acer us boo payed ina Nev cain nce ae "Te ules, state and fat ofthe thre games ave s00g sina’ studied Nevada baccarat with William E, Walden Smt tssvely because te casos weve splayed wore ear I} accesible, Our technigs can be cried ove othe ober foams of bacart and chemin de fe "ie were oiginaly motte by the oeration hat baccarat and chen de er have several pots of resemblance fo the game St blackjack, or The fat that practical wining Stateice fv twenty-one have been discovered sugested that fhetnigh doe praca winning ttl fr baccarat and hen defer. incor ote sian intwent-one, we found tare are no caret pracal wing statis forthe main past of th games efor the money Ranker and Player bts Rules and Procedures ‘To besin the Nevada baccarat game, cight decks of cards are shuffled and a joker is placed face upnear the end, tbe cards are then pu into a wooden dealing box calle a shoe. The first cards ‘exposed, and is values noted, face cards being counted as tens. ‘Then this uuniber of cards is discarded, or *bumed.” "The table has twelve seats, occupied by an assortment of customers and sil sili house employee who bets money fand pretends to be a player in order to attract customers oF ‘Smale play. We refer to them indiscriminately a “player.” ‘There are (wo principal bets, called “Banker” and “Players.” ‘Anyplayer may make either ofthese bets before the beginning of fay round of play, oF “coup.” To begin the eveing’s pay, two ofthe players are singled out, ‘Oneis termed The Banker and he otheris termed The Player. The seats are numbered counterclockwise from one totwehe. Player 0 Bocce ‘number one i initally The Banker, unless he refuses. In thiscase the opportunity ate vounterlockwise aroun the table wn someone accepts. The Players generally chosen tobe that player, other than The Banker, wo has the lngest bet on the Player. We hhave not noticed an oecasion when there were no bets on The Player. When we played, there were shils in the game and they ‘generally bet on The Plver (except when acting as The Banker, ‘when they generally bet on The Bankes). “The Banker fais the shoe and deals as long as the bet “Banker” (which we also refer to as a bet on The Banker) does not lose, When the bet “Players (which we aso refer to asa bet ‘on The Player) wins, the shoe moves tothe player on the right. ‘This player now becomes The Banker. I the coup isa tie, the players are allowed to alter their bes in any manner they wish. ‘The same Banker then deals another coup. To begin a coup, The Danker and The Player are dealt (wo ‘cards each. As we oted above, the cards ace through nine are each worth her fae valueand tens and face cards ar each worth zero points. Only the las distin the total is counted. ‘After The Banker and The Player each receive two cards, the ‘roupier faces their hands. IF ether wwo-card total equals 8 or 9 ((ermed anatural 8 ora natural 9, asthe case may be), all bets are sealed at once. If nether The Player nor The Banker have a natural, The layer and The Banker then draw orstand according tothe set of| rules in Table 3-1. ‘The high hand wins. Ifthe hands are equal, thereisateand no money changeshands. Playersare the freeto change their betsin any desired manner. If the coup being played is complete when tie joker is reached, the shoe cnds and the cards azereshuled. (Otherwise the coup is ist played out to completion. Then the shoe ends and the card are reshuffled. However, the casino may reshulle the cards at anytime between coups. The Mathematis of Gambling Table 341 Player having 05 draws acard 67 stands 89 tums cards over Banker having draws whan does not dkaw when Tho Playercraws The Player draws o one, 0.8 1 one, 09 2 ‘one, 08 3 ‘none, 07,9 8 4 one, 27 0.1.89 5 one, 47 0389 6 67 one, 05, 8,9 7 stands stands 8 tums cards over tune cards ovor 8 tumscards over turns cards over ‘The Main Bets ‘Two main bets agsinst the house can be made. One can bet on ther The Banker or The Player. Winning bets on The Player are paidat even money. Winning beison The Banker arepaid0.95of fhe armour Ux. The five percent tax which i imposed on what otherwise would have been an evenmoney pay-off is called “igorish.” For eight complete decks, the probability that The ‘Baanker wins is 0.458597, andthe probabily of a ies 0.095156. “The basic idea of thecaculation ofthese numbers isto consider all possible distinc sit-card sequences. The outcome for each sequence is computed and the corresponding probability of that 2 acca sequence is computed and aocumulated in the appropriate egter. Numerous shortcuts, which simplify and abbreviate the calculation, were taken. ‘The house advantage (we use advantage as a synonym for mathematical expectation) over The Player is 1.2351 percent. The house advantage over The Banker is 0.458597X5 percent—1.2351 percent or 1.0579 percent, where 2.2930 percent {sthe effectiveousetax on The Banker's winnings. Itiesarenot counted as trials, then the figures for house advantage should be ‘ultplied by 1/0.904844, which give a house advantage per bet that isnot ate, over The Banker of 1.1692 percent and over The layer of 1.3680 percent. The effective house taxon The Banker in this stuation is 2.5341 percent. ‘Weattemptedto determine whether or not the abnormal com- positions of the shoe, which arise as successive coups ae dealt, sve rise to uctuations in the expectations of The Banker and “The Player bets which are sufficient to overcome the house ede. [tturs out that this oocasionaly happens but the fluctuations are not large enough nor frequent enough to be the basis ofa practical ‘winning strategy. This was determined in two ways. Fist, we varied the quantity of cards of a single numerical value. The results were negative. "Wenextinguired as to whether, ifone were able to analyze the endeeck perfectly (eg. the player might receive radioed instruc- ‘tions froma compute) there were appreciable player advantages con ether bot aignificnt part ofthe time. We selected 29 xs of 13 cards each, each set drawn randomly from eight complete decks. There'were small positive expectations in only two instanoss out of $8, Once The Player had a 3.2% edge and once ‘The Banker had a0.1% edge. ‘Wenext proved, byargumentstoo lengthy andintricateto give Ihre, that the probability distributions desribing the conditional expectations of The Banker and The Player epread out as the ‘number of unplayed cards decreases. Thus there are fewer advan- tageous bets of each type, and they are less advantageous, a the numberof unplayed cards increases above 13. The converse oc- Ps The Mathematics of Gambling (0182 1101 Yoed uy siUjed Os 44 0} Zt WHO Peonpas KERB. ° e eo0- - = sooo- e- e000~ z- ue t100- (Panupuor) 7-¢ 21%, t e- 2000- ats no0+ so00+ 00+ sroo0+ wwzovo+ qunog weg (UexUeA)sONIeA tag Joye erewpoxdyuy —woderpejey yo Be uENY (peg qw8ia) euE0rS Te TR coor scoo+ soot hoo+ 2000+ zioo- 2000- sooo wnoo- ‘we000- yea s0Keig uy e6ubuD en eo enren jo PIED ‘euo yoes1qns The Mathematics of Ganbling curs asthe numberof unplayed cards decreased below 13. "The observed practical minimum ranged from eight 0 17 in ‘one casino and from 20 up in another. The theoretical minimum, ‘when no cards are burned, is six. Ths the rests for 13 unplayed ards seem 10 conclusively demonstrate that no practical win- ‘ing strategy is possible forthe Nevada game, even with a com puting machine playing a perfect game. ‘To see why, consider te accompanying Table 3-2 (pp. 4-35), based on Table 2 of Walden's thesis rom this Table we see the effect of removing one of any card from the eight-deck baccarat pack. Proceeding in the way we developed the theory of Blackjack, we get relative pont vas ‘which ae listed in the next to the last column. The last column ves a simpler approximate point count system. ‘We would now like 10 know how powerful a point count system n baccarat s compared with poi count systems in black- Jack. Todo this we compute theroot mean square (RMS) value of the column called “Change in Advantage of Banker Bet." "We do this by equaring each of those numbers, counting the square for zero-value cards four times because there are four times as many. Then we add these square, divide by 13 and take the square root. The resulting root mean square or RMS valueis O64, That measures now lat the deck shifts fromis base ta- ing value fora full pack. “Taking one card ofa given rank (we think of there being 13 sanks) changes the fraction of ay ofthese 18 ranks by anamonnt 532/416-31/415 which equals .00222. If we divide the RMS value by this vale we get (288 asa measure of how rapidly thisadvan- tage ofthe two bes shifts from the starting value asthe compas: tion ofthe deck changes. 'Now we ze going to compare this with the situation in back jack, Table 23, for one-deck blackjack, can be treated in the Ses way to aus how fart the advantage changes in blackjack. ‘The Table is from Peter Griffin's book, Theory of Blackjack Revised, page 44, We get RMS valu of 0.467%. Thecorrespond- ing change in the fraction of asinglerank when one cardi drawn, 6 Baccarat Table 33 (One-Deck Blackjack Subtract One Card of Value | Change In Advantage of Payer -081% 035% oases 055% 083% 046% 028% 00% eovroanane 18% 51% {34/52/51 or O18. The rao of RMS valu o this value is ‘258, If we dvi this byte corresponding re in boa cezi’rwhhcn tester counting as changin plajer advantage or ddvanage shifts in ines as fast in blackjack as it does in baccarat. ee a The Mathematics of Gambling [Note that dividing the RMS value by the “change in fraction” ‘ofa single peck adjusts the one deck blackjack figures and the ‘Gehtdleck baccarat figures so they are comparable. If we had ted, for example, an eight dock blackjack table instead, we still ‘would have had a final rato of about nine times. “This allows us to translate how well a point count in baccarat ‘works compared with one in blackjack. ln baccarat we start Out ‘wih more thana Ie disadvantage and with eight decks, Imagine {blackjack game with an eight deck pack and a 18 disadvan. tage, Now imagine play continues through the blackjack deck. “The blackjack deck advantage froma ~ 1%e starting level might, ‘on very rare oceasions, shift 9% fo a + 8% advantage. ‘As often as this happenin blackjack would be the approx imate frequency with which we would get 1/9th as much shift in baccarat; meaning froma ~ 1Mfadvantageto a OMe advantage or break even forthe banker bet, Since there are two bets, banker and player, the player Bet would also be break even or beter about as often. ‘The conclusionis that you night expect to break even or better ineight dock baccarat about twice as often a you woul expectto have an BM edge in eight-deck blackjack. How often would you have a Io advantage in eight deck baccarat? About twice as ‘oficnasyou would gota 7M edgein blackjack. Theobvimsean- ‘luson that advantagesin baccarat are very small, they are very fare and the few that occur are measly always in the lst five to 20 cards inthe pack. ‘The Tie Bet In addition to wagers onthe Player or Banker hands, the casinos sank and Player hands Ihave the seme total, this bet gains nine times the amount bet. Otherwise the bet i lost. The probability ofa tie is 9.51565, heace the expectation ofthe bet is ~4884%. it is lear, however, thatthe probability and thus the expecta- tion of a tie depends on the subset of unplayed cards. For in- stance, in theexireme ad improbableevent thatthe residual deck ” consist solely often valuecards, the probability ofatieisequlto ‘ne and the expectation nine Thit card cing Sete Potent advantagous. ‘Using computer smulton, random subs of diferent sizes vere selected from a complete l6card (ight deck) pack. The Fesuls were dsappotning fiom a money-making pespec= tivemthe advaniages which oocur with complete knowledge of ‘he wed cards ae limited tothe extreme end ofthe pack ad ate generally not large. Practical card counting rates are at bet ‘rg, and at bes precarious, forthe are easy elminated by shuffling the deck with 25 cards remaining.