CARD COUNTING FUNDAMENTALS Perhaps the best way to protect your games against the card counter, is to learn to think like a professional player. Only then will you come away with a true, understanding and appreciation for the skills necessary to play this game on a serious level. Only then, will you know what it takes to win. It's timeV-to look at the mechanics of playing the game skillfully. Over the last forty years, the most significant change in the technical aspects of card counting can be summed up in one word: simplification. Learning to count cards is easier today than ever; before. Fundamentally, there are five steps in the process: basic strategy, counting, true count, playing, and betting strategies. Forgive me for the rehash, but X will fly through these fundamentals in the fewest words possible, yet attempt to share with you some perspective and insight not generally found in books on card counting. Basic Strategy Learning to count cards starts with basic strategy, the core of all winning strategies. Simply stated, it's the best way to play any two cards against the dealer s upcard when no other information is taken into consideration. The strategy changes slightly depending on the number of decks and rules. We've assumed multideck as our base strategy; single-deck changes are noted. If your club offers double alter split, or surrender, make sure to learn the cotrect plays for these rules. ^: So, you would hit 12 versus the dealers 2,?^dIff2l^rsus|3astam^^faU otherj'stiffs against a small card, and nit allistiffs'against a big card^^S^^^^^^my^pf; better. Stiffs, which represent almost 3^%fo|ElTn^ds^ar^l5ginig»Rrop^itipns;^un^m^pcafd.;; When you hit 16;vHsus" 7, for example, the goal isto^^^i^yot^ loss." Your 16 versus the Healers 7 only wins when the dealer breaks, abouW6%1of^ hands. Tnemetloss is~48 bets'(74 - 26 = 48). Hitting wins about 30% for a net loss of 40 bets (^0;- 30 = .40). Although both piaysjj^e^hitpng Joses^ess, only 40 bets .versus, 48:;Ur: Basic ^strategy is all about .comparing?yo’^options and making the aecisipnftna| loseplessg ^wins more, or^ occasionally turns a losing hand into %Aymnerli These simple rules, when added to^the blackjacks, and totals of 5,6, 7, 8,17,18,19,20, A-8 or A-9, represent about 75.% of the complete basic strategy. There are alfew; fihefpphts'rega^ (marked with an asterisk in the previous table). In single-deck game^double'll versus an; ace/;9 versus 2,{8 versus 5, and 8>versus’6.;. |p§iThe mathematics of doubling down are worth mentioning. It’s ldgical&fass'ume that by doubling ,:down and’restricting the hand toonejhituardonl^ybujmust lessen the probability of winning. This jil!S;true..Ybulosejmof e'hands:by doubling! But there’s a*catchv,;For example, when doublirig lO versus] 7, hitting wins 63% of the hands, a net; profit oT26'-bsfa^(63|ia^7^=-26'hl!clbubluteJftiB!^^SD^B^ net'(profitfofi20 bets (60,^- 40 = 20). The double down winners, however, are double bets. Assuming $100 units, hitting wins $2,600 (26 x $ 100)>■' doubling wins $4,QQo|||C) x $200). Hitting wins more vh'andsl doubling wins more rnoney it’s that simple. it Adding the hard doubles, about 12.5%rofiaU hands', brings us' to almost 90% of the complete strategy. he fine points of basic strategy allow more aggressive soft'doubiing in single-deck. ‘Double A-2 and A-3 versus 4, and ^-6,yersus 2. You may even double A-8 versus 6 if die dealer 'stands on all 17s. You can also double A^-7 versus ;2imtwopr more decks if the dealer hits soft 17.

There aren't any fine points foripair, splitting, but therefis'a common rule variation generally seen in shoe games that affects the.strategy and that's the double-after-split. When offered, basic str^egy^allows^more aggressive splil^Split 2-2 and 3-3 versus 2fi2r2 and 3-3 versus 3> and 6p^e^us 2. Splitfo.-6 versus 7 in single- and double-deck; games. Even 4-4 versus 5, and 4-4 versus 6 can now SefpitvSiii V - The other ruletmdstioqmmonly offeredHs late surrender, and it, to^pnan^sihel!strategy.4The imostimportantpnangesaresurrenderl6jyer^us:;9,16 versus:10M6versusrace;and 15.versus 10. ■' Finally/-never vtkke insurance. Not 'eve'n with blackjack, which is a common misconception amongst many gamers. The mathematics just; don’t fly. Take a six-deck shoe, and your A-10 versus an ace. There are 214non-tens;i resulting im^p^,bf 214 insurance bets^and there are 95 ten-yalues| remaining, resulting in blackjafcl«fpr the dealer, ^yith the'2 to;Tparoffi'insurahce wins 190lbet|| 190). The net loss is 24 bets$214 —For every 309m(adev(214 + 95jjj on^^mge,'this There youlwf[j||baSic|n?jj5tljpmhe^fs,ffi is counting, tmcking7er^lowr^^n&rjadvantage techniques, or cheating, basic strategy is still used roi play the nSjority of all hands. ; If you/desi^mbr^etau^^^^^^^^^n^etelDasic strategy [for anyhumber of decks in ■ iM^rjfo^BlWl^ck. Thi^^^k'alsbKpjrovKleaitKr^prrect composition-depehdent|play^ ■ multiple Sra^ombinations, sucnra^caTOW^mp^fVersus'dOlf^^^ is comprised of Most plays'-areapplicable t®sihgle^deck;|^‘;i Basic Strategy s Worth Basic strategy wasmrstSewom6di|^^ ’apprb^fo^^^ _were-'^|ed^ Computer simulations helped- refine the^rategy, buti they were subject to 16 versus 5, a computer would deal millions, of hands,-'first standing, then hitting, then comparing the results. For this particular hand, the results afe’decisive; standing waaclearlyifhe;<rtpct decision by\ a huge margin, and no^atistlcd:^^^^^lc^Mnge !tlfe|'tesidt||A p^djlike,j^er^ 4, how<^»/M|o| close to a coin flip that, even after one! million simulated hands/you^ill 'coidd^ 100% sure.^|; Today, sophisticated computer/prbgrams’'are|now cycle .through every possible player hand combination, dealer hand com!bihatibn, and drawing se<juence‘fo \york out the exact strategy. . . As a base game, assume single^deck, the dealer, stands on all 17s,and you can double ,on|any: two cards. There are 55 different player hands and 10 possible dealer up cards for a total pf,550 possible player/dealer combinations. What happens when you work out the bestplay for each ha^ up jthe irespectivb?gainsfaridilosses?; Professor. Peter Griffin says it best: !, he remarkable thihg'is that the bottom line, or netresultofth'efemr^am^^^'turns out'tcrbe zero' when rounded off to the nearest tenth%f& 'game Just think about this. For many decades, unbeknownst to the industry, this popular game with the help of a few simple rules was;«fp^all^practical purposes, a dead even gamble! Rules Numerous rules have been evaluated for the game, ranging from re-split aces,^double three|orSdre cards, to a sixthrjseven-card no-bust automatic win, but there are only a handful of rules that encompass most major^mMphs. They aretirnultiple decks, hit'soft l^fpouble on 10 and litonly,'jdpuble afte| jsplit, late surrender, and 6 to 5 blackjack bonuses. SdiherUiles are good for the house, others are^^Sj i;for theplayer. Lets take a look, and remember, all of the following percentages are^appjoximate. Two or more decks ?j$Vmentioned above, single-deck gaihifWanicJ-'om . essentially a dead even game, -Tor’ two ^decks, the casinQ^g^j^^fflq^fe^5Sa;^,'‘ for the shoe game of four to eight decks, the casino/eage’ is approximately .5%. ^ Hit Soft 17,, This rule costs the basic strateg^lave^ao^^^^b^i [•' Double^.djpnd 11 Only; ! The norm in Reno and Tahoe^ these restrictions costthe basic strategy player about .2%. KDouble After Split The double after split option is worth about,0.14% to the basic strategy;player??--1’ Late ^Surrender The late surrender option is;worth about 0.1% to the basic strategy player. Pays 6 to 5 pii Blackjack £> This rule costs the playelr, about E4%. v l^fn the blackjack world, rules are expressed in percentages, As an example, a two-deck deal costs theplayer 0.3%, which is an additional 0.3% of all money put into action (the original wagers). Double-after-split is a favorable rule to the player by about 0.14%. ^rh^aiit^^^^'pkyerLMns1 an additional 0.14%;bf alkmbjney initially bet with these rules. i:. One Syay to get a better feel for what these numbers are worth is tcTfirst convert them to?a decimal number by, simply moving the decimal point two places toThe. left, and, assume a coin flipping game, For 0.i%f; the decimalmumber is';.003,^r3bets for|everyiL000 flips. So,pn^cQuld say that playing against double-deck costs the basic strategy player about three additional bets in every 1,000 hands, The logic is not perfect due to doubles, splits, blackjack; bonuses? and the fact that blackjack is not a 50/50 proposition, but it will help yon put these' smalls percentages into perspectives The Public s Play The casinp advantage in blackjackhs'an individual advantage, as it changes with the skill of each player. In-1987?a very important studyjtitea’Maflemof^lEx^ecf® was conducted by|TProf^^r Peter Griffin, He quantified how w^lltn^pubiic played Blackjack group. Griffin made over 11,000 observations of individual players;m'Reno, Tahoe, vegas^na'jNew Jersey. Taking a small sample of playing decisions from eacnfhe assessed the cost ofi^^^uSmistkkes, put them all together, arid concluded that the average blackjack player loses at a rate jusri:ufld^I2|0^ (includes the 0.5% casino starting advantage). gtturiously/New Jersey players rated better tnari those frompjasIVegas, which Gfiffiri^attributed c^nSi^eptionally^fast dealersonautomatic pilot, who ; inadvertently assisted th^e pl^^rspvith/a betteri^yate^^^^tl^^Falrlolfix^aceu^hd^atnesB'sj that the players tend to school one another. Weak pla^^^qmckl^leMftlro^laylpetteKfdUowirig the typical; sarcasm, criticism, and outright ridicule froniotl^^la^r^'i^^ciallyitrue in jurisdictions where the, same players play together every, day%The st'udy <’v^silater published iffiGriffin^^p^tri^^^^Imai ^Ramblings in 1991, a!ndlc^^miavmlabl^f|you want tofcnecklou^he details^pjj Basic Strategy and The Industry Basic strategy^rafigrmrmousrimportarice to the mai^^^S^defrdm^beirig't^threshrildlbyi^^c^J we gauge the smUfofftnepl^^wm^E^apeS»tept^^ts|^efo@ne^6s‘t^igriifica^fa^o^^ecnhgr'|;| hold percentages. If the theoretical edge|ag^^^he ;typical^^^^^^Sund f rihe, <’ basic strategy player? for eyeryf^^MMpla^K wn5l^S^pasil?^:i^e^y;th^^^^igpret ical win ^ drops by TawNtt Without Basic Stfategy, > 10 hours X;;100 hands/hour. x ■'$l'odM.yerag^feet^ |||With Basic Strategy lb hours x 100 hands/hour x $100 : average J}eP-xlQ.'5% casino'edge If, for any reason, the public woke up one day arid decided to platlbasicfstrategy flie least: of the industry’s problems would be card counters, trackers,,arid^ne^ersaHQlH|p'etQ@^cs|^ffluld take a nosedive and the, potential loss of revenue wouldhe staggering. What’^particularly disconcerting the realization that jumping frorn typical player to basic strategy player is remarkably, easy and takes little time;—most could learn the strategy while having lunch. CARD COUNTING SYSTEMS ;Tne"fjurpose ofany^iiM^oAnfing'-strategy-is to provide the player with an ’bihgbirig^u^^the remaining cards. The counter is trying-to identify;situations where the^pjninfda^sare^ii^irit high cards, or more highcaMs than probability would indicate.1 In these sij^iftdhs,'^epj^yar benefits in many ways: more blackjacks (only the player gets paid 3 to 2), more dealer busts (dealer must draw to. 16 or less), and the player options of doubling, splitting, and insurance become moreprofitable (rrotl available'to the dealer). Systems fall into two broad categories; balanced and unBalanced. Balanced Counts The Hi^Lpjw^is^popular balanced ®rd counting system with ijtwenty small'cards (2^3/4,‘ 5,^d%)l counted as !4T;■fyvbrt^big^cards (10^p^!^andj|^l ciojtnted as -|i>* and the neutral cards (7,8, and 9) I valued at zero. The card counter adds ohefor^^^small card played, subtracts one for every big card -played, and ment^yimkmtiuhs. .Epuntti'-*'. equal number of high and low*cards, balanced counts must end bn zero arm^llfhe cards are played. If a count assigns values of-1,,;0, and +l,’itis called a one-level coundv’^VEbn the ace is assigned a count value of zero, it is generally side-counted for, betting purposes.

few who have played with fiye-level counts!1, With^tw^fo^v^e tiers tq^thk count/yalues/ systems are known as multilevel or 'advar^ed point also called APGIIIm "Tne multilevel Thorp count has b'eeh'pegged the ultimate count/and' it exists as a tneor?tical' exercise only. Unbalanced Counts p^nl^lfabedp^ountpysremsVaregaininffiinTOo^laMn^Mariv;.beli^fetnata‘few years {from now'allbut the diehardpl^^tq^ the'pa^(M^lj^ploynn^i^^|Pqri|3^^1agagm^^9<^^CtyQimt,. there .^e.twenty-fe>unsm|Ufca^l^^3;!4,"5^^^^^^^^^^Pl;and!jqnljpwent^blffi^^^lO/>LiP>K, 'and A) counted as -1. Un^^^^M^mntsj do fnnCei^^m^^^^fter’all' cards are played, unless you i^start your count atjsome^pbintotherjdfSiizeaij^Hj Which Count is Best? Which count is best? The Ten-Count, Hi-Low, Hi-Opt II, Reveres APC, the Knockout Count? There are numerous choices, and most of the early Research on ,card counting was bogged down trying® answer this question. Thanks to forty years of computer simulations, we now know, regardless of complexity, mat] most card rountingsptems produce, similarwin rates. In some games, mosdy deeply dealt pitch games, a multilevel count may outperform a ohe-level couht 'Dy ohe- or two-tenths of a percent, butmot always. To help understand why/ assume1 ajsixrdeck game, three, deck's^emain and the running'count is +3. How many'possible three-Ldecbjc6mbinati6tt?sTocftB3.^ist?lBillions? !I;really;doht have a clue, but what I do know, is that there are sfymany that iom^ttrers/littlejwhether you count the five as a<+l or +2, or whether you only count the three, four, five, and six as small cards, or add the deurce, or add the seven, and so on. One is looking for precision when none exists. Its a lot like counting the ripples in the ocean. You start from the right and I'll start from the left, which strategy is more accurate? So, it should come as no surprise, the card counting strategy you choose is not a crucial factor. They're all about the same, and they all do a reasonable job of evaluating an incredibly complex game. How Far Can We Go? Many, years back I was sweating my mentor, Big JB, one of the very best players of all time. He picked up A-7 and stood against a ten. This was puzzling since we didn't even have a deviation for the play. It had to be a mistake, but this guy is a machine and never makes mistakes. Later, I asked him about the play. He quickly explained, "There were no threes left, only one deuce, and one ace. Where was I going with the hand?" "He really didn't know if the play was right or wrong, he was just playing instinctively. I was startled. It was the first time that I realized that the count may not always indicate the correct play. I later discovered that these composition-dependent plays were first Worked out by Peter Griffin, later published by David Sklansky in "Gambling Times" in 1977, and then in Getting the Best of It in • 1982 under the title, “The Key-Card Concept". 4 Whereas a counting system such as the Hi-Low looks?at the game by grouping cards into small, big, and neutral categories, some plays are impacted more by counting a single value. For example, you're halfway down in the single-deck game, you have 15 versus 9, so what card impacts your hand the most? The five or six is the most common answer, but neither, is correct. When the deck is rich in',. ■ fives and sixes you have a conflict; ^hitting appears to help your hand, but you may also want to stand since the dealer's holecard is also more likely to be a'five or six, resulting in totals of 14 or, 15:; Now look at the impact of the seven. For every extra seven remaining, you lean towards standing as seven '3! will bust the hand just as easily as a ten. Moreover, extra sevens put the dealer on 16, the weakest possible total, so you definitely want to stand. No matter how you look at it, standing is correct, there is no conflict. The seven impacts the hand more than any other card. So, occasionally, advantages may be derived from this kind of detail, provided you're able to to side count all cards. Many years ago I developed a system for single-deck called Total Recall Scanning (TRS). It enabled me to count all cards (ten parameters), and at any time know exactly what's left. It employed a mechanical method bf counting versus the use of mnemonics (memory aids). Some of the cards were counted mentally, others with different parts of the body, and some were not counted, just noted. On a practical level, TRS proved to be more of a.stunt than a sophisticated card counting system, but counting all cards is far from impossible. Learning to utilize the information accurately and profitably, however, is extemely difficult. - Although some players may keep one or more side counts in. addition to their primary count, side counting aces being the most common, the advantage from all this extra effort is very, very small.

Stanford Wong once said that the i use of a more complicated counting system (versus the high-low) is primarily ego gratification." , /.-Another interesting side note is that Wong once hailed Peter Griffin as the best counter of all. Some readers suggested otherwise. He then pointed out that Griffin counted six parameters: a primary count, four side b^^^auud total number ofi^^layed cards. 1. Hi-Opt I with 2^3; 4,5 as +1, and 10, J, Q, K 2. Aces -g^3MSevensw- 4:'-Eights 5. Nines 6. Total cards unplayed,-veSusmsingTa true ra^^gB |pLfe|went om to-say, "If I find-someone wh^^^mts more than six things (wM^^ a’computer) v-and knows how to use tmpmormaHoh, then I will retract my statement. . .’’ Wong also pointed out thefdistinction'between the be^^oi^er^nd tne best player, and that is that the best player ends up with the^^rej^| Which Count for Gamers? My-advice is to learri the j side counts1^^S?it|has<' beemaround fofsollong, andv^Suatealbv^^m^^^^^Ss^^^^ffl^^cbrifident-ith^tlt^w^^^^l {fstrategy changes are accurate. Its^^OThe^^^^o^^^pimt^among players^''■ a - The best way ^Spractice the'Hi-^^^^^kart withfone card at a time,-th^yle^^tjie two^caHl combinations to emulate liyetpla^^^^^^^^^^^^^t’down a deckunfabout 3(^^^^^^^uyefgot ; it; To test your accuracy, placepjnyxard face down,’ count the remaining 51 cards/and predict the value of that unknown card. jlf your count is -lllmelast card-will be a small card to^getkHe’count back to zero; if your count is +1, the last card jWill be a big ca:rd‘f c);-g^the^^^tt back to zero; and if the count is'-zero,- the last carai will1 be a 7, 8, or 9. : ' You may even check out an unbalanced,count|as you’re likely; loplcirig at a trend that figures to be around for as long as blackjack survives. How Counts Look and Play Most card counting'Strategies evaluate the game similarly, so when astrategyds'played according to the book (no c^ouflage)^most&Qunts look and play the same. There is^ojyeve’r,{the occasional discrepancy. For hand held games, multilevel counts tend to react quicker; that is, they tend to find profitable situations faster. Here’s an example thafo^paresxhe Hi-Low to Reveres APC (the 14 count). Assume a single-deck game, and if the first ten cards played are 5,4,10, 5,10, 3,7,4, A, and 10, then an APC player has a count of+8 and the Hi-Low players count is only +1. One can easily see how an APC player might move his money faster, or deviate from basic play sooner than a Hi-Low player. Another situation may arise with insurance. Point counts are not particularly strong for assessing insurance decisions. When the dealer shows an ace, the only relevant information needed to make the best decision is the ratio of ten-values to non-tens, so the best count for this particular play is the Ten-Count, not a point count. To illustrate a shortcoming of the point count, consider single-deck. The first ten cards played are 2,10,10, 7, 8, 5,10, A, 9, A, and the player picks up 8-9 versus an ace. The Hi-Low count is -4, the APC is -6; should either player take insurance? Not according to these counts, but the best players might. A closer look shows that 3 ten-values and 10 non-tens have been played. This leaves 13 ten-values and 26 non-tens for a ratio of exactly 2 to 1. Since insurance pays 2 to 1, the true odds based on the remaining cards, insurance is a dead even bet. Some pros may decide to minimize their fluctuation with a good hand or simply take insurance for cover. Yet, what's more revealing, look what happens when we add one more small card. Although neither count changes enough to indicate taking insurance, taking insurance is the correct play! The good player knows the limitations of his count. The most practical way for players to add this accuracy to the game is by playing two-handed. One player uses the primary count or point count, the other plays a Ten-Count for insurance decisions only. Its a common ploy with couples. If this seems like too much work, you should know that insurance is the most important decision in the game, so some players will go to great lengths to pick up the extra accuracy. An amusing situation often arises when two bosses, each using a different count, get different plays and want to know who is right. Well, they may both be right! Obviously, only one play is correct in the mathematical sense, and, given the complexities of the game, only a computer knows the right answer. But, if both counts are played accurately, they're both considered right. The next time you're sweating a player, and he happens to move his money too fast, or deviates from basic strategy too soon, or takes insurance when the count indicates otherwise, or just doesn’t seem to be doing anything according to your count, the player may still have a mathematically justifiable reason! THE TRUE COUNT Counting cards starts with the running count, your cumulative count from adding the small cards (+1) and subtracting the big darcl^(^|p At^any time,’the running count may!bfe worth rriwe^r less ■ depending on the number of decks remaininjg^A. +10 with five decks left/is*hot the same as *+10 with only one deck left. The fewer the number of decks remaining, the richer the coMcewhatioij bf high cards, and, therefore, the better jpl'fdf the player. Notice that with five decks left, ten extra high cards can be spread out. One would only expect about two extra high cards per deck. With only one deck and ten extra high cards/rvow we re talkihg abodran extra high card every few cards bn average, which figures to have a real impact on how you play and bet. In essence, the true countjindicates what to expect, one deck at a time. With one deck left, the true count is five; times as strong than with five decks left> four times as strong than with four decks left, and so on. One needs an adjustment that takes the Inumber of decks remaining into consideration,^ana that adjustment is called the 'true count'. To get a true count, divide the running coimt'by the, number of decks remaining. With five decks left, a running count of +10ls’'dividea'by five decks to, get a+ru^ left, a running count of + 10 is divided by two'^ets a true +5. The pnly.time the rurining count equals the true countis when orie declqren^n^smce^e^e^iviaing by one. The true countfis used^ith1 alllbalan^d poin^cbunt’s^ and with some systems/the true icouiit may be based on half-decks.?. ^Kpnverting a running count to a true co^^^^^whe^onebstimates the remaining cards in full deck increments. But what happens when the remaining qeqks are better described as ;2Vi decks or 3% decks? To.quickly divide the runnihg^cbuht by fractions canjbe-daunting. It’s for these reasons that many gamers learn to couridc^ds^but ayojdjlear tq^wqrki^ith true counts becausfotheyi find the forjmcjsi: players, too. There are many tricks and shortcuts thatb^U dramatically simplify the process. As an example, wirh fractional; estimates/of thec^cksyer^^ both the .running(counCand the number of decks remaining be^re^yi(^^.'\/fej^hjg.-^^^ofl2idivided;by,^^|diwdcs is ,equal to a running count of 24 divided by 3, both having a true count(of;+8. A running count of 14 divided by 3% decks is equal to a running count of 28 divided by 7, or a +4 true count. Just double-double, .when' necessary, and go horn there. . Another method, and arguably the best‘o£|^is division completely, as you can get the same answer with multiplication! Take a rurining count of + 10 with five decks remaining. One can either divide by 5, or multiply by .2 to get the same true +2; The .2 is a special number called a conversion factor'. These are never calculated during play, they are worked out beforehand by simply dividing 1 by the number of decks remaining. With four decks remaining, the conversion factor is 1/4= .25'. With 2Vi decks remaining, the conversion factor is 1/2,5 = .4. If the number of cards remaining is> less than one deck, say 17; cards, divide 1 by the fraction 17/ 52 to get a conversion factor of 3. Now, here’s where'we pun it all together, and I suSpect yd®|will be surprised at howlsimple this process really is. Next is chart for six decks; .ae^t^^.^lh the last column I have simply rounded off the

Great stuff! I assume the documents were translated or something? There are some errors in it. Is there much more to add?

I'm joking about "hidden translations" but the wording of the stuff is just funny, because it wasn't bothered to be fixed. Example: So, you would hit 12 versus the dealers 2,?^dIff2l^rsus|3astam^^faU otherj'stiffs against a small card, and nit allistiffs'against a big card^^S^^^^^^my^pf; better.

Sorry about that. It's the text reader not working right. I'll see what I can do to avoid the problem.

Cards games are not really my thing but a player can force the draw of a card to benefit another player.