An application of expert information to win betting on the Kentucky Derby, 1981–2005

Abstract The Kentucky Derby features top three-year-old thoroughbred horses. Run at miles, it is typically at least 1/8 mile longer than any of the horses has raced before. This extra distance, usually combined with a large field, makes the race a difficult test of stamina for horses this young. Bettors, because there is no direct evidence of whether a horse has the stamina to compete effectively at miles, are also challenged. The informational content of one publicly available, pedigree-based measure of stamina, the Dosage Index, is used with simple performance measures to identify a semi-strong-form inefficiency, and to create a betting scheme based on the optimal capital growth model that merges these criteria with the public’s opinion. Statistically significant profits, net of transaction costs, could have been achieved during the period 1981 to 2005.