The Favorite-Longshot Midas

64 Pages Posted: 14 Nov 2018 Last revised: 12 Aug 2020

See all articles by Etan Green

Etan Green

Wharton - Operations, Information and Decisions

Haksoo Lee

Stanford Law School

David M. Rothschild

Microsoft Research

Date Written: February 11, 2020


Common valuations pose an obstacle to trade and thus an existential threat to brokers, who profit from taxing trade. We write down a model in which a broker drives a wedge between valuations by deceiving gullible traders. Separate valuations facilitate arbitrage, and arbitrage generates brokerage fees. We then show that this scheme explains a classic case of market inefficiency: the favorite-longshot bias in horserace parimutuel markets. Racetracks induce naive bettors to overbet longshots by promulgating predictions that overestimate the chances of longshots. This creates arbitrage opportunities for sophisticated bettors and additional income for tracks. We document substantial cross-track variation in the extent of the favorite-longshot bias. Using original data, we show that this variation is predicted by the degree to which tracks overestimate the chances of longshots. Previous explanations for the favorite-longshot bias—i.e., speculation, probability weighting, and belief heterogeneity—can only explain cross-track variation in the bias with preferences or beliefs that vary idiosyncratically by track.

Keywords: deception, arbitrage, parallax and tax, favorite-longshot bias, neural network, structural estimation

JEL Classification: D22, D82, D83, L83

Suggested Citation

Green, Etan and Lee, Haksoo and Rothschild, David M., The Favorite-Longshot Midas (February 11, 2020). Jacobs Levy Equity Management Center for Quantitative Financial Research Paper, Available at SSRN: or

Etan Green (Contact Author)

Wharton - Operations, Information and Decisions ( email )

Philadelphia, PA 19104
United States

Haksoo Lee

Stanford Law School ( email )

559 Nathan Abbott Way
Stanford, CA 94305-8610
United States

David M. Rothschild

Microsoft Research ( email )

Redomond, WA 98052

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics