Efficiency and the Disposition Effect in NFL Prediction Markets by Jack Vogel, Ph.D., Alpha Architect
- Hartzmark and Solomon
- A version of the paper can be found here.
- Want a summary of academic papers with alpha? Check out our Academic Research Recap Category.
Examining NFL betting contracts at Tradesports.com, we find mispricing consistent with the disposition effect, where investors are more likely to close out profitable positions than losing positions. Prices are too low when teams are ahead and too high when teams are behind. Returns following news events exhibit short-term reversals and longer-term momentum. These results do not appear driven by liquidity or non-financial reasons for trade. Finding the disposition effect in a negative expected return gambling market questions standard explanations for the effect (belief in mean reversion, prospect theory). It is consistent with cognitive dissonance, and models with time-inconsistent behavior.
This paper examines gambling behavior on Tradesports.com for the 2003-2004, 2004-2005, and 2005-2006 NFL seasons. The main test is the following: “the contract price at any time should represent an unbiased estimate of the true probability of the event occurring.”
So are gambling markets efficient?
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.
So next time you are in Vegas and decide to place a bet, remember this study!
If you liked this post, don’t forget to subscribe to Alpha Architect