Back | Contents | Next Chapter 24
* * * * * * * * * * * * * * * * * * * *
How to win at blackjack
Soon after I had learnt how to memorize playing cards, it occurred to me that there must be a way of cashing in on my new found ability. Blackjack seemed like a natural target. It involved skill (unlike roulette or dice, which are based on pure chance), and I was already familiar with the game. I also felt there was a score to be settled: I had lost many more times than I had ever won!
I had always thought that beating the bank was a romantic but ill-conceived notion - the stuff of fiction and a sure-fire way of losing even more money. It might be possible in a Graham Greene novel, but never in real life. Memorizing thirty-five decks of cards put a different complexion on things. Today, I am barred from casinos all over Britain and France. One or two will let me in for a drink, but if I get anywhere near the blackjack tables, I am back out on the street. They know that I have devised a winning strategy, and if I played for long enough, I could break the bank. I don't want to encourage anyone to take up gambling - there are many other ways of making money - but my approach to blackjack is a good example of what can be achieved with a trained memory. THE GAMEThe object of blackjack is for the player to be dealt cards that add up to 21, or as close as possible, without 'busting'. The opposing dealer must draw cards totalling a minimum of 17. Whoever is closest to 21 wins that particular hand. The skill, for the player, lies in deciding how many cards he or she should draw, relative to the degree of risk. As is my nature (my stubborn streak again), I wanted to work out whether it was possible to gain an edge over the dealer. I proceeded to deal myself thousands of hands, analysing every possible permutation. After six months, I had studied 100,000 hands. I never intended to deal so many cards, but once I had started, I was overcome with a relentless urge to continue playing and amassing results. The only way to test theories satisfactorily was to carry out thousands of individual trials. KEEPING IT IN THE FAMILYYou may find the thought of devoting so much time to a card game abhorrent, or at least a trifle excessive. I often wondered at the time what was really keeping me going. I think I now know, and it is quite uncanny. After I had carried out all these experiments, I came across a 1932 newspaper article about the game of bridge. In December of that year, the London Evening Standard published a series of five articles by Dr E. Gordon Reeve on the 'Reeveu' system for auction and contract bridge, invented by Gordon himself. In the article, he says the following: Three years of illness gave me the opportunity to work out the possibilities of scoring game. I dealt 5,000 hands, and each hand was played by all four players - North, South, East and West, in all the denominations respectively. Thus, the results of 100,000 combinations of hands were tabulated. It was a strange feeling coming across such a precedent; it was also comforting to know that I wasn't the only person fanatical enough to be lured into the monotonous world of card permutations. But imagine the shiver that went down my spine when 1 discovered that this man, whom I had never met (he died in 1938), was in fact my grandfather. CARD-COUNTINGOne of the first discoveries I made during my experiments was realizing that I would usually win if low cards had been removed from the deck. Conversely, if high cards (10s, court cards, and aces) had been removed, the bank won the majority of hands. By keeping a constant check or tally on which cards had been dealt, I was able to judge, at any stage during the game, whether or not the conditions were favourable. If they were good (lots of low cards removed), I would stake large bets; if they were poor (lots of high removed), I would place the minimum bet. This strategy is known as 'card-counting1. Card-counters are rife throughout the casino world. They are the scourge of club managers, even though they are not doing anything illegal. Most of them are small-time gamblers who nibble away at clubs' profits. They never win large amounts, but they still annoy the management. If they are spotted (most tables these days are monitored by sophisticated closed-circuit TV), they are usually asked to leave, and politely told never to darken the doors again. (Casinos are private clubs, allowing the management to reserve the right of entry or to rescind membership.) Known card-counters are also likely to feature in the Griffin Book, a three-volume tome compiled by a Las Vegas detective agency. It is circulated worldwide among casino managers, and lists a variety of undesirables, everyone from trouble-makers to card-counters. I have never seen a copy, but I gather it includes photographs, stills taken from the security cameras. HIGH ROLLERSSet apart from the hoi polloi of small-time card-counters are a handful of supreme professionals, or 'high rollers'. They can make upwards of £500,000 tax-free, annually. Utterly dedicated to their work, these are the card-counting elite. They operate either on their own or in small groups, and are virtually impossible to identify. They are always on the move, flying from one country to the next, constantly changing their identities and adopting a variety of disguises. Most of them are American or Canadian. Two are based in England. One, known as 'the Professor', lives in the Midlands and has been known to dress as a woman. The other, alas, has been forced to hand in his chips. STAYING ON THE RIGHT SIDE OF THE LAW (OF PROBABILITIES)After dealing 100,000 hands, I felt I had got to know the heart and soul of blackjack. Every aspect of the game had been dissected and held up to the light. I had developed a basic card-counting strategy to the point where the bank's overall advantage was reduced to a half of one per cent. In other words, for every ?100 that I bet during the game, I would be returned £99.50, providing my stake remained constant ('flat betting'). If, however, I substantially increased my bet when the cards were favourable, I could realize a profit of £1 to £2 for every ?100 of turnover staked. This might not sound a lot, but it soon adds up. If your initial stake is £100, for example, you can turn over £10,000 in an evening. It was time to put theory into practice. I began by joining as many clubs as I could, all over the country. Profits were modest to begin with, but there were other perks of the job. I embarked on a pleasant tour of the casinos along the south coast, enjoying what I call 'free evenings': my profit would cover the cost of travel, meals, and drink. It wasn't long before I was targetting the Midlands and certain London clubs, returning home every morning with a reasonable profit. The strategy was working. More important, the casino managers appeared to be tolerating my presence. I began to earn a good living, about £500 to £600 per week, and I was learning to ride the ups and downs. I remember getting off to a particularly bad start on my first visit to a club in the Midlands. Within half an hour, I was £500 down. I decided that a good dinner was in order. After dining on a sumptuous steak, washed down with a delightful wine, I was pleasantly surprised to find that my dinner bill had been 'taken care of by the manager. He had spotted a punter with potential. Managers do this from time to time, to encourage you to gamble even more money. I returned to my blackjack table, whereupon I not only recouped my losses but ended up showing a profit of £500. I tried to share my delight with the manager, celebrating my change of fortune and thanking him for the delicious dinner. The look on his face signalled the beginning of the end of a beautiful friendship. After two more similar visits, I was barred. It is hard to describe the thrill of placing heavy bets in a casino, especially a glamorous one, knowing that you have a clear advantage over the bank. But there were downsides to my chosen career. It's exhausting having to look over your shoulder all the time, waiting for the manager's discreet words in your ear, 'Mr O'Brien, could you come with me please.' (It wasn't always that polite.) After a while, I was no longer satisfied with my earnings. It was small reward for a dangerous, itinerant lifestyle. I yearned for more and more profit and was soon taking home ?1,000 per day. It was then that I became a marked man. Word travels fast in the casino world. Scores of letters began to drop through the letter box, terminating my membership of casinos nationwide. 'Dear Mr O'Brien,' read one from a club in Luton, 'it has been decided at an extraordinary meeting of the Election Committee that your membership be withdrawn with immediate effect. This means that you will no longer be allowed to visit the club either as a member or as a guest.' Many people think it is unfair to bar a player who merely beats a casino at its own game, particularly when there is nothing more than mental skill involved. I was doing nothing illegal. But I can understand the casino's point of view: they are in the business of making money, so why should they tolerate someone who reduces their profit margins? Besides, if I am barred, it is my own fault for making myself conspicuous in the first place. BACK TO THE DRAWING BOARDI was convinced that I was being barred because of my betting strategy. Most of the time, I would stake the minimum permitted amount (usually £5). When I calculated a clear advantage, however, I would raise it to £25, £50, or £100. Increasing it by a factor of twenty inevitably attracted the attention of the casino inspectors, but it was the only way I could capitalize on the odds. Or so I thought. Back at the drawing board, I read all the best books on the game and managed to acquire a print-out from Las Vegas listing thousands of possible hands and what to do in each situation. Using a computer, I proved and disproved every existing theory I could lay my hands on. This rime, however, I was able to deal millions of hands in a matter of hours, thanks to the computer. I finally arrived at an optimum strategy for winning, which I plan to publish in its entirety in a separate book. It requires a trained memory, a cool nerve, and simple mental arithmetic. I will, however, disclose a few details now, to give you an idea of how it works. It's one of life's little ironies that I am no longer able to use it myself, although I did test it out recently on a lucrative tour of France's casinos (where my face was unfamiliar), but more of that later. THE MOMENT OF TRUTHAs I said earlier, the card-counter's skill is to predict which cards are left in the shoe. People do this in a variety of ways, some more subtle than others. My approach, a variation on existing methods, is to assign a very specific value to each card as it is dealt. A high card has a minus value, and a low card has a plus value. (They range from approximately -2 to +2.) As the shoe progresses, I keep a running total of the overall value, which I divide by a figure (anything between 1 and 8) reflecting the number of cards still to be dealt. This gives me what is known as a 'true count'. In blackjack, you arc required to place your bet before the cards have been dealt. If, after the previous hand, the true count is greater than +.75, I will increase my bet for the next hand: the laws of probability tell me that the concentration of high cards still in the shoe has increased. If the true count drops below +.75, I know that there is a greater concentration of low cards still to be dealt. High cards, remember, give me an advantage. Low cards give the dealer an advantage. Let me explain a little more about the number that I use to divide the overall value of the cards. In Britain, one shoe of cards consists of four decks. The dealer will place a blank card somewhere near the end of the shoe. This is known as the 'cut', and it is where the dealing stops. Card-counters prefer the cut to be as close to the natural end of the shoe as possible, for reasons that will become apparent. At the beginning of the shoe, I divide the overall value by 8. Let's assume the game has just started and only five cards have been dealt. They are all low cards and the overall value is +6. It would be foolish to conclude from a mere five cards that a high card is likely to follow, which is why I divide the value by so much. The true count then becomes +.75 (8 divided by 6) and I don't increase my bet. As the number of cards left in the shoe decreases, I divide by 7, then by 6, then by 5, and so on. In other words, the true count is calculated in proportion to the amount of cards remaining. (In France, where casinos play with six decks, I initially divide the pack by 12.) Unless the croupier is inexperienced, you are unlikely to find yourself dividing by 1. The cut usually comes first. For the sake of example, though, let's assume that it is a very good cut and there are only a few cards left in the shoe. A lot of low cards have been dealt, so many, in fact, that the overall value is +12. I would divide this figure by 1, still leaving me with a true count of +12 (an advantageous situation to be in). This means that there is a high concentration of court cards left in the few remaining cards still to be dealt. I increase my bet accordingly. Using such a finely calibrated 'true count' allows me to adopt a more inconspicuous betting strategy. All I need now is a good disguise. TRACKINGMy strategy incorporates many other technical features, most of which will not mean much to the uninitiated. 'Ace tracking', 'count tracking,' and 'sequence tracking', for example, can all be mastered with a trained memory. Sometimes an inexperienced croupier won't shuffle a shoe thoroughly. Imagine the advantage you would suddenly have if you had memorized sequences of cards from the previous shoe (a technique you learnt in Chapter 16 when you memorized one deck of cards). UNORTHODOX CALLSThere are times when knowledge of the true count can lead to some very unusual calls. For example, let's suppose that my first two cards add up to 12, 13, 14, 15, or 16. The dealer's card, which is always face up, is a 2, 3, 4, 5, or 6. I know that the true count is -6; in other words, there are a lot of low cards left in the shoe. It's not a good situation for me, and the dealer is likely to win. Most players would stick. Knowing that I am likely to be dealt a low card, however, I break with tradition and ask for more cards. Supposing the dealer has a 6 and I am on 13. I ask for another card, 5 say, and then stick on 18. The dealer takes a card, 6, and another, 5, making 17. I have won. If I had stayed on 13, however, the dealer would have drawn a 5, then a 6, making 17. I have lost. It doesn't always work like this, of course, but it's a way of making the best of a bad situation. Whenever I make strange calls, it always amuses me to hear the accusatory comments and criticisms from other players at the table. 'You're obviously knew to this game, aren't you?' or 'Take my advice, if you want to win, never make a call like that.' Some people get quite upset and start claiming that my unorthodox calls are the cause of their ill-fortune. STRATEGY NOT SYSTEMI don't consider myself a gambler. I play to a strategy not a system. Over the last few years, the face of the compulsive gambler has become an all too familiar sight. I see them with their own 'winning' systems, some of which work for a while, but they never make money in the long term. That is why casinos love them - they are a bread-and-butter source of income. The strategy player is the complete antithesis. I may lose occasionally, but the underlying trend is always upwards. The only chance I have had to demonstrate my revised strategy was in the Autumn of 1992, when GQ, magazine arranged for me to play the casinos of northern France. The four-day trip was based on the assumption that I was an unknown quantity in France. I played at seven casinos and won in six of them. Using my new strategy, I was able to bet more subtly, gradually increasing and decreasing my stake. In five memorable hours at the Grand Casino in Dieppe, I made £1,200, much to the annoyance of the management, who were beside themselves. Once again, they were kind enough to pay for my meal, after which I cashed in my chips and headed for the casino at Deauville. It all ended dramatically in Enghien les Bains, a casino in the northern suburbs of Paris. It was my last day of the trip and I had turned my original float of £4,000 into £6,000. I had been playing for only twenty minutes, when the manager tapped me on the shoulder and uttered those immortal words. 'Mr O'Brien? We must ask you to leave immediately.1 It was not my method of play that betrayed me. They had calculated from my geographical movements that I was professional player. Why would someone staying at a hotel in Deauville travel to Paris to play blackjack? THE ROLE OF MEMORYIt is not very easy to adopt my strategy without a trained memory. On a simple level, your overall concentration and powers of observation are so much sharper if you have worked on improving your memory. They need to be: I am often sitting at the table for five or six hours without a break. And in today's casinos, you are being scrutinized from every possible angle. I have quite often found myself playing with three security cameras trained on my table, a croupier watching my every move and an inspector looking over my shoulder! Most card-counters arc easy to spot. They give themselves away by covering their mouths with their hands, trying to conceal lip movements as they frantically struggle to keep count of the cards. They scan the cards with conspicuous head and eye movements, and their play is characterized by long pauses between cards. I have trained my memory, concentration and observation to the point where I can keep pace with the fastest of dealers, hold a conversation with the inspector and make spontaneous calculations at the same time. I once overheard a croupier in Dieppe observe to a passing inspector 'Il est trop machine'. This was an apt description, as I was working robotically. Memory also plays a vital role when I have to refer back to a mental reference grid that I have compiled. Using location, I can access the print-out from Las Vegas, the books I have read, and my own statistical findings. It is a vast data base, equipping me for every possible hand. For example, let's assume my first two cards total 12, I immediately refer to a location based around the Peacock Theatre in Woking. (12 - AB. My own person for AB is Alan Bennett, the actor and playwright). The dealer's card is a 2, which tells me to locate the second stage along the journey: the box office. I have a mental image of bars across the ticket window and the man inside wearing handcuffs, as if he was in a prison. Handcuffs gives me a coded true count of +3. I now know not to draw any more cards if the true count equals or exceeds this level. They are likely to be high, and I could go bust. All guesswork has been completely eliminated from my game. I know there is an optimum decision for every situation, enabling me to act like a robot rather than a gambler. Next time you visit a casino, look at the man playing blackjack on his own. Look closer still. It might be me!