Investigation of the Matching Hypothesis

This article uses panel data on Major League Baseball managers to perform a direct test of the matching hypothesis. This study uses measures of manager and team performance that are more efficient than those previously used to test the matching hypothesis in the context of Major League Baseball. The study also uses a new variable that proxies the effect manager-team matches have on output. Using these variables allows a direct test of the matching hypothesis. Estimation of both the fixed effect and random effects models reveal that the quality of a manager-team match is an important determinant of production.