Professor Lloyd Cohen, of George Mason University, posted some of his observations on gambling psychology on the Econlaw discussion list. I asked if I could post them, and he not only did, but sent me lengthier notes on blackjack. Here are paragraphs making three good points, on the choice of house percentages, on moral flaws that make people lose, and on the practical application of the Gambler's Ruin paradox:

Casino managers, in general, are no better schooled in the science of probability and statistics than are typical blackjack players. The games and rules casinos choose to offer the public, rather than being the product of deductive mathematics, merely represent what has worked successfully for them over tens of thousands of trials. They employ hundreds of dealers twentyfour hours a day and thus have an enormous sample from which to infer the relevant characteristics of the blackjack probability distribution. They are, in general, interested in asking only one question, at what rate will we make money. The optimal rate for them is the highest one that will not drive away the patrons. Historically, blackjack has been one of the most profitable casino games. The house wins approximately 6% of each dollar bet, about double their win rate at craps.

...

The most common form of playing error is bottomed on cowardice, bizarrely combined and compounded by another character flawa love of gambling. The thrill of gambling requires risking an amount of money that represents a meaningful loss. Once the money is risked, players often freeze; they prefer docile inaction to action. Thus the most frequent form of error is to fail to take appropriate risks when the situations are favorable. For example, when the opportunity to double or split arises the typical player often holds back because he fears risking the extra money. Similarly when sitting with a sixteen facing the dealer's eight the player will often refuse to take a hit. At the same time the symmetrical error of hitting when it is not called for is only rarely made. While it may be that this can be explained by a desire to sustain the tension for a longer period of time, my guess is that its true root is simple cowardice. One is willing to die, but not by one's own hand.

...

Odd though it might seem, ordinary gamblers sometimes have larger and more satisfying wins at the tables than successful card counters. This results from the money management required of a skilled blackjack player. The successful card counter knows that he will lose almost as often as he wins. Hence he can not afford to be frivolous with his winnings, they must be available to pay off his losses. In addition, because the card counter's advantage is less than 2% he must limit the percentage of his bankroll that he risks on each bet in order to ensure that he does not fall victim to `gambler's ruin', i.e., the tendency of gamblers', even those with an advantage over the house, to bet too large a percentage of their bankroll and end up broke when the eventual bad run of cards materializes. A successful player with a networth of say $200,000 should feel comfortable risking no more than $10,000 on a three day weekend trip. Further, prudence dictates that in order not to fall victim to gambler's ruin he must probably limit his bets to approximately $200 per hand.

To return to Eric Rasmusen's weblog, click http://php.indiana.edu/~erasmuse/w/0.rasmusen.htm.