Reference Point Roulette13. April 2012 by Joachim Goldberg
I can easily understand why someone would buy a lottery ticket in the slim hope of getting rich in a single coup. Even though the expected return from this investment is negative, so no economically- rational person should do it, millions of people take the plunge every week. The reason is because the very low probability of hitting the jackpot is likely to be perceived as greater than it really is. In addition, the thing one buys with a lottery ticket, above all else, is the right to dream.
In contrast, the reason people go into a casino in the hope of winning at the roulette table has remained an enduring puzzle for me. Although the chance of winning on any single spin of the wheel is significantly better than in a national lottery, the expected return is still negative. There is no chance of becoming immensely wealthy from playing just once and repeated plays only increase the certainty that one will have less money on leaving the table than on joining it. Now there are certainly some out there who claim to have developed a system that permits long-run profits in casino games, like roulette. But most people do not possess such a system and yet, despite their aversion to losses, they continue to visit the casino. There is obviously no quick route to profits in roulette. Being paid out 35 times one’s stake with a bet on a single number is fun, but one will have to commit a good deal of money to the gamble before wealth beckons. Perhaps it is the casino ambience that players find so appealing: the piles of plastic chips; the envious onlookers; the champagne and lobster in the hotel restaurant after the big win.
Social scientists have investigated this seemingly puzzling behaviour. It is well documented that people are generally unwilling to accept a gamble where the probability of winning or losing a given cash amount is identical. Despite this, casinos are full of people who are willing to gamble on red or black, over and over again. One would be correct in arguing that, for small amounts, loss aversion simply does not hold. Low-value stakes, for example, one euro, are negligible compared to the typical player’s income and so fall below the threshold for perception. The player is effectively indifferent to potential losses when the stakes are low. It is precisely this effect that casinos exploit when they want to encourage players to play at the table. Casinos deliberately set high table limits, not because they do not want to run the risk of large loss, but because it serves as a reference point for the players; a high table limit makes the player’s stake appear small and therefore less perceptible. The goal, of course, is to increase the player’s willingness to join in. Once, engaged, and especially after a series of small losses, the gambler is caught in the trap. Thereafter, the desire to make good the losses, typically with a higher stake, will take over as the principal motivation. At that point, the reference point will cease to be the table limit; it will be the gambler’s pocketbook.
 Simmons, Joseph P., Novemsky, Nathan: From Loss Aversion to Loss Acceptance: Context Effects on Loss Aversion in Risky Choice (2009)