Abstract

We use a unique data set from Finnish and Swedish horse race betting markets to explain the favorite-longshot bias. The data set includes a complete set of odds for exotic markets. We use the exotic market odds in conjunction with the win market odds and find convincing support for the misperceptions explanation of the favorite-longshot bias rather than the risk-love explanation. Furthermore, our data provide evidence of a specific type of failure to reduce compound lotteries. Namely, it seems that bettors do not assess the exotic market events as simple lotteries but instead consider the race for the first place and the race for the second place in a sequential form.