Probability of Winning in Risky Choices

Media reports say that high earners and syndicates buy lottery tickets in bulk. Experimental evidence shows that agents aggressively bid in auctions and contests. Do people try to trade-off the probability of winning with other basic risk dimensions (for example, cost) to achieve a subjective threshold probability of winning (in environments they can) even when such choices are second-order stochastic dominated? The literature on risky choices suggests so. In the main design of this experiment, we deconstruct the expected value with variance and skewness of a lottery with Bernoulli distribution to examine the decision-making process. Based on the results, a proportion is classified as expected utility maximizer (EUM) while another proportion seems to achieve a subjective threshold probability of winning (termed as target probability of winning (TPW)). More TPWs prefer higher probabilities compared to EUMs in a constant value lottery set which may explain the preference for negative skewness in experiments. Additionally, we test two contest designs and find TPWs in the population which may explain the puzzle of equilibrium effort more than risk-neutral Nash equilibrium in experiments