Gambling and The Self: A Sure Bet (To Lose Money)
On the psychological appeal and trappings of gambling.
Posted May 29, 2012
Gambling in the United States is a rapidly growing industry. Currently, 20 states permit commercialized gaming and another 18 allow native american groups to offer gambling instead. As gaming-related revenues exceed $100B annually (American Gaming Association, 2012), many states are turning to gambling to attract tourism and to generate tax revenues ($8B of taxes in 2011). In my own state of Ohio, several new commercial casinos have recently opened or are soon to open their doors in the near future.
Although the economic appeal of gambling is clear, the increased availability of gambling (e.g., more states with casinos, internet wagering) leads one to ponder its appeal psychologically. It is an undeniable truth that nearly all forms of gambling will, in the long run, guarantee losing money. In the face of this reality, why do people flock to casinos to bet well over $60B annually?
Undoubtedly, there is no simple answer to this question. Gambling certainly serves a constellation of psychological needs, and for any given individual, the qualities that make gambling attractive will vary. Yet, when examining the research literature, there are some themes that offer some answers to this question. In this blog, I focus on just a few that implicate the self.
Erroneous perceptions of control
If people examined most gaming activities in terms of raw statistics (e.g., lotteries, slot machines), the only sure bet is losing money. Yet, people routinely perceive games as "winnable" and "controllable" activities despite the fact that they are random by design and engineered to be money-losing.
For example, consider state lotteries. In my state of Ohio as well as many other states, the basic lottery involves picking 6 numbers out of 49, and the odds of winning the big jackpot (i.e., picking all six numbers correctly) is 1 in 14 million (13,983,816 to be precise). Rather than "winning the whole jackpot," one could get 5 of the 6 numbers correct and win $1500 (the odds of that are merely 1 in 54,201). When examining these odds, buying a lottery ticket seems ridiculous (after all, according to the National Weather Service, the odds of being struck by lightning in one's lifetime is 1 in 10,000—yes, one is five times less likely to even win "the runner-up prize" of $1500 in the lottery than to be struck by lightning). Yet, lotteries generate considerable post-expense revenue (over $18B in the USA, including $675M in Ohio alone).
Why do people bet when the likelihood of losing money is so incredibly high? One factor that matters is that games of chance are presented as games of skill. Consider the language involved—people "play the lottery," which makes it sound like one is playing a sport or a musical instrument rather than participating in a randomly-determined gamble. However, the odds are about as "naked" as possible (e.g., six ping-pong balls fall out of a machine with 49 in it), yet the game is framed as people "picking numbers." Because the draw is randomly determined, one is just as likely to win by letting a lottery computer pick their numbers for them, but because people get to pick numbers (and in some cases, get to pick "special numbers" like "the powerball"), people's perceptions of control are artificially inflated when they are actually nonexistent.
This phenomenon was shown in an elegant series of series by Langer (1975). In one experiment, she had workers either pick a lottery ticket or they were randomly given a lottery ticket—each ticket cost $1 and the winning pot was $50. Later, she had another person come along and offer to buy the ticket, allowing the researcher to determine the "market price" of one's lottery ticket. Those who got to pick their ticket wanted $8.67 to sell their $1 ticket, whereas those who were randomly given their $1 ticket only asked for $1.96. Despite the fact that the odds of winning were exactly the same, people who got to "pick their ticket" wanted more than four times as much money to sell their ticket, illustrating the illusion of control in a totally random gamble.
Although a lottery provides an opportunity to see the nakedness of this fallacy, other games obscure the randomness considerably. For example, consider the game of craps where a player (called the shooter) rolls two dice and everyone at the table bets on the outcome of each roll. For any roll (odds bets excluded), the average house edge ranges from 1.36% to nearly 17%. A number of the more popular bets (e.g., 4 the hard way, C&E bets, Yo bets) are especially bad bets and yield loss rates in excess of 11% per roll. However, the odds become obscured by the dizzying range of choices a bettor has for each roll, and thus people routinely make financially less sound bets from the array of options before them. In addition to myriad choices, there is a vocabulary and array of rules that one must learn (e.g., when certain bets can and cannot be made), and suddenly, the illusion of skill grows as people have "to master" a system that is designed to obscure the pure randomness of the situation.
While discussing craps, I should mention another classic study that illustrated how people have illusions of control. In a classic study by Goffman (1967), they observed that when craps shooters needed a larger number to win (e.g., having two dice add up to 10), shooters threw the dice harder than when shooters needed a smaller number to win (e.g., having two dice add up to 4). In other words, shooters implicitly believed that harder throws would produce bigger numbers and softer throws would produce smaller numbers—but clearly, the force with which dice are thrown has no effect on the number obtained.
If this seems difficult to believe, think about all of the superstitious behaviors one observes in casinos (e.g., people blowing on dice before throwing them, people not picking up their cards in blackjack until after the dealer picks up the house's cards)—the only way that superstitious behaviors can seem beneficial is if one believes (erroneously, obviously) that these acts can affect gaming outcomes.
Overly optimistic self-beliefs
In addition to believing that one has more control over random events than is warranted, it is also the case that people have unrealistically positive illusions about the self. A common finding in the literature is that the vast majority of people believe that they are better than average—they will be happier in life, be less likely to experience divorce, live longer, and be less likely to develop diseases than the average person (Armor & Taylor, 2002; Weinstein, 1980).
Clearly, the majority of people cannot be better than average, but this unrealistic optimism also can make people feel like they are more likely to win in gambling than the average person. Other people are the ones who lose in gambling, but one can believe that they will beat the odds because of this powerful self illusion. Moreover, our memories are often biased in self-serving ways. That is, we're more likely to remember the times we won and less likely to remember the times we lost (and how much money we lost). As a result, gamblers both bring unrealistic control and unrealistic optimism with them to the casino—and in the end, they are likely to leave the casino with less money because of it.
Poor statistical skills and reasoning
Of course, all of the above presupposes that people can assess and determine odds correctly. However, it turns out that people's statistical ability is typically quite poor. For example, although most people know that the odds of rolling any given number with one die is 1 in 6 (approximately 17%) few people understand how to compute the odds for more complicated events.
For example, the likelihood of rolling 2 particular numbers with two dice, such as a 6-6, is less than 3%. Most people, sadly including many college graduates who have taken statistics courses, neither understand the statistical rule conceptually (it's the mathematical product of the two probabilities, or .1667 x 2) nor do they have a good intuitive feel for them. When gambling outcomes become more complex than simple one event outcomes, people's weaknesses in understanding odds and probabilities can become quite expensive.
Just as an example, in the game of craps, one can bet that any given roll will be a 12 (two 6s), and if this outcome occurs, the payout is quite handsome (i.e., one will win $30 for each dollar bet). People who are drawn to "the big win" will often place this bet. Yet, as just noted above, people greatly overestimate the likelihood of this 1-in-36 event (which is why the house has a 14% edge on this particular bet).
Features of casinos impair gamblers' judgment
As noted in the previous section, people often do not know how to calculate the correct odds. And, even if they do, they must be able to effortfully render calculations to determine the odds. Yet, casinos are not places designed to give people the best conditions to employ their cognitive horsepower. Instead, they are noisy, distracting places, which reduce people's cognitive resources and increases environmental distractions. Further, in places like Nevada, patrons are served free alcoholic beverages, which further impair gamblers' ability to render accurate assessments of their game play.
Other aspects of casino games themselves provide gamblers with nondiagnostic cues that affect betting behavior. For example, in the game of roulette, a wheel is spun and a ball is dropped onto the wheel, which eventually comes to rest in 1 of 38 spots (0, 00, or 1-36). People place a variety of bets, such as on a particular number or a color as the outcome of each spin. Each wheel spin is a unique event, and thus any outcome has the same likelihood of occurrence on each spin. However, one fixture of roulette tables are display boards that show the results of the last several spins.
Objectively, this information is completely irrelevant, yet people use it to inform their betting. Ironically, two different people could look at a particular display and come to completely opposing conclusions (e.g., "36 has come up a few times, it's hot so I should bet it" vs. "5 hasn't come up in a long time, it's due"), and each interpretation is erroneous because each spin is a unique event and not dependent on a previous outcome. The casino provides this information, however, to encourage betting. Regardless of any given bettor's specious beliefs, the casino makes more money when people bet more, and providing people with useless information promotes it.
Individual differences in gambling susceptibility
In addition to environments providing people with distractions that undercut gamblers' performance, there are also personality characteristics that predict more problematic gambling. For example, one study involving over 1000 Canadian adolescents found that those who were diagnosed with ADHD (attention deficit hyperactivity disorder) were more likely to gamble and more likely to develop gambling problems (Faregh & Derevensky, 2010). Thus, it seems that having difficulty in being able to control one's impulsive reactions can make gambling more appealing and betting more problematic.
Another study that examined gambling behavior from a large database of twins (over 4000 pairs) in Australia (Slutske et al., 2009) found evidence of a moderate effect of genetic components in predicting 11 different forms of gambling. In another study by Slutske and her colleagues (2012), they found that children categorized at age 3 as having a temperament characterized by behavioral and emotional undercontrol were twice as likely to show problems with gambling in adulthood (ages 21-32), and these differences could not be explained by IQ or socioeconomic status. Although it would be too simplistic to conclude that people with impulse control problems are more likely to reveal disordered gambling, it is likely that impulsivity coupled with other factors (e.g., anxiety, negative emotions) increase the likelihood of gambling problems.
The above analysis presents just a few of the factors that contribute to the appeal of gambling and why people are inclined to wager money that, in all likelihood, will be lost. Gambling can certainly provide people with a form of entertainment (e.g., no one expects to make money when they go to dinner or attend a music concert), and certainly many people routinely make trips to Vegas or to nearby race tracks without having a "gambling problem." However, it is also clear that people often do not approach gaming in a rational, logical fashion, and that self illusions are often at the center of why this occurs. As gambling becomes more widespread and socially sanctioned, understanding the factors that contribute to problematic gambling behavior is increasingly important.