Do Market Participants Misprice Lottery-Type Assets? Evidence from the European Soccer Betting Market

This paper addresses findings from previous asset pricing research that reveal lottery-like stocks are mispriced. This conclusion, however, relies on asset pricing models which might suffer from the joint hypothesis problem: That is, abnormal returns can reflect market inefficiencies, a bad asset pricing model or both. I contribute to the discussion by examining lottery-type assets on the European betting market. Compared to stock markets, betting markets offer the advantage that no asset pricing model is necessary to test market efficiency because outcomes are observable at a terminal point. I use a unique data set of soccer odds covering both a betting exchange and the bookmaker market. The findings reveal a favorite-longshot bias in both market structures confirming the findings from previous literature that lottery-type assets are mispriced. An expected utility model and a prospect theory model confirm that the favorite-longshot bias on the betting exchange is due to misperception of probabilities rather than risk-love. Although bookmakers also bias odds there is evidence that bookmakers are rational in protecting themselves from adverse selection and/or in increasing their turnover. This conclusion is supported by further analysis on the tennis betting market