Structural Breaks in the Game: The Case of Major League Baseball
Published online on July 06, 2015
Abstract
To search for eras in a sports league, we utilize time-series tests with structural breaks in Major League Baseball performance. Using data from 1871-2010, the mean and standard deviation of four different performance measures are examined. Throughout, rather than assume that a break point is known a priori, we identify breaks endogenously from the data. Perhaps most notable among our findings, we identify a deterministic trend in mean slugging percentage with breaks in 1921 and 1992. Interestingly, these years closely coincide with the early years of the free-swinging (Babe Ruth) era and the modern steroid era, respectively.