Download This Paper Open PDF in Browser

A Natural Experiment for Efficient Markets: Information Quality, Behavioral Measurements and Influential Agents

49 Pages Posted: 14 Feb 2018 Last revised: 20 Feb 2018

Brian Mills

University of Florida

Steven Salaga

University of Georgia

Date Written: February 2, 2018

Abstract

We test the integration of repeated decision making behavior of influential agents in asset prices, using Major League Baseball as a unique empirical setting. Our approach exploits a natural experiment in which umpire assignments are revealed only for certain games, and makes use of more than 2.5 million decisions made by these game officials that have influence over game outcomes. Estimations reveal that the totals market only partially adjusts to information related to heterogeneity in umpire decision-making tendencies. We show that this information is exploitable by informed bettors, providing advantages over more salient, lower quality information related to umpire own-lagged game outcomes. These results suggest that underlying individual decision information making can serve as a high quality indicator of asset values due to its persistence across time.

Keywords: Market Efficiency, Betting Markets, Behavior, Baseball

JEL Classification: G14, L83, Z2

Suggested Citation

Mills, Brian and Salaga, Steven, A Natural Experiment for Efficient Markets: Information Quality, Behavioral Measurements and Influential Agents (February 2, 2018). Available at SSRN: https://ssrn.com/abstract=3112783

Brian Mills (Contact Author)

University of Florida ( email )

PO Box 117165, 201 Stuzin Hall
Gainesville, FL 32610-0496
United States

Steven Salaga

University of Georgia ( email )

330 River Road
361 Ramsey
Athens, GA 30602
United States
7062541173 (Phone)

Paper statistics

Downloads
30
Abstract Views
166
PlumX