The hubbub surrounding the proposed Harrahs' casino in Rhode Island got me pondering an old question. Why do people gamble? I'm not talking about friendly bets that merely add excitement to sports games and the Oscars. Why do people gamble at casinos, where every game is stacked against them? Don't they understand that this is a bad idea - that the odds can't be beaten in the long run?
Of course, some people simply don't understand the idea of probability. Nevertheless, most gamblers know that probabilistic laws reign supreme, and yet they gamble anyway. The possibility of winning simply carries more weight in their decision process than the probability of losing. It's irrational.
There's no reason to believe that gamblers behave differently inside a casino than they do elsewhere. Gamblers are probably living their whole lives in accordance with this same kind of reasoning. From an evolutionary perspective, the universality of gambling indicates that this trait hasn't been selected against, and this begs the question: What advantage could possibly be conferred by recklessly ignoring probabilistic laws?
My theory is that the basis for gambling behavior - irrational confidence - is a liability when dealing with physical phenomena, like decks of cards, but confers an advantage in social situations. As a corollary, it also confers an advantage in sexual selection and is therefore evolutionarily favorable.
A 1999 paper by economists Eddie Dekel and Suzanne Scotchmer reaches a similar conclusion. It demonstrates that risk-taking behavior is favored in some winner-take-all games (like patent races). For many species, winner-take-all games determine a male's ability to reproduce. This hypothesis is supported by the fact that more males than females gamble.
That is one piece of the puzzle. But I think the advantages of a gambler's logic are actually much broader than that. The reason is that in social situations involving incomplete information, each party has to assess the confidence of the other and use it as a proxy for missing information.
Take the following example. Two strangers are considering whether to fight over a prize. Neither party is willing to fight if he thinks he'll probably lose. The first thing they do is size each other up. The information gained from this process is incomplete: They can see that they are of equal size, but neither party knows whether the other is a karate master. Let's say that neither actually is, but that the first party is a rationalist, and the second is a gambler. The rationalist will see his odds as 50-50 if he believes the other guy is not a karate master. However, if he observes confidence in his opponent (which might suggest karate skills), he will logically adjust his own chances downward and surrender the prize. The second party (the gambler) would have mirrored that thought process if he were rational, but he hasn't. His confidence is not linked to a rational probability assessment, and that allows him to pull off the perfect bluff.
A job interview is another good example. Other than assessing an applicant's social compatibility, the primary purpose of most interviews is to determine the applicant's confidence. A resumé only gives partial information - confidence serves as the proxy for all the unknowns. Applicants are therefore more likely to get hired if their confidence is not anchored to a rational estimation of how good they'll be at the job.
It is also worth noting that elected leaders - who bluff and maneuver for a living, and whom we select using confidence as a huge proxy - often display a weird level of confidence in their actions.
Gamblers have definite advantages in competitive social situations because other people react to their confidence and thereby increase the gambler's real chances. Irrational confidence only gets you into trouble when it causes you to engage in chest-puffing matches against physical phenomena (like roulette wheels, or people twice your size). So go ahead and look down your nose at the people wasting their money at the slots. Just don't be surprised when they take your job. And your girlfriend.
Matt Prewitt '08 thinks he can beat the odds at Harrah's.
