Abstract
A structural model is derived describing the relationship between the probability that the favored team is victorious and the point spread through the use of a no-arbitrage condition. A series of probit models are developed which show that wagering markets were efficient within season between 1988-2006 and that the point spread is a statistically-significant indicator of probability of victory. A team-level excess performance metric is estimated, termed NFL Alphas, which indicates the extent to which a team's on-the-field performance exceeded betting market expectations. Finally, it is shown that there exists a market over-reaction effect for Alphas across seasons.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
