Empirical Determination of Baseball Eras: Multivariate Changepoint Analysis in Major League Baseball

TL;DR

This study applies multivariate change point analysis to a century of baseball data to empirically identify distinct eras, revealing significant shifts in performance and strategy that align with historical periods and external influences.

Contribution

It introduces a multivariate change point methodology to detect structural shifts in baseball history, providing more robust era identification than previous univariate approaches.

Findings

Confirmed known baseball eras such as Dead Ball and Steroid Era

Identified periods of team dynasties and collapses

Demonstrated the effectiveness of multivariate analysis in sports history

Abstract

We use multivariate change point analysis methods, to identify not only mean shifts but also changes in variance across a wide array of statistical time series. Our primary objective is to empirically discern distinct eras in the evolution of baseball, shedding light on significant transformations in team performance and management strategies. We leverage a rich dataset comprising baseball statistics from the late 1800s to 2020, spanning over a century of the sport's history. Results confirm previous historical research, pinpointing well-known baseball eras, such as the Dead Ball Era, Integration Era, Steroid Era, and Post-Steroid Era. Moreover, the study delves into the detection of substantial changes in team performance, effectively identifying periods of both dynasties and collapses within a team's history. The multivariate change point analysis proves to be a valuable tool for understanding the intricate dynamics of baseball's evolution. The method offers a data-driven approach to unveil structural shifts in the sport's historical landscape, providing fresh insights into the impact of rule changes, player strategies, and external factors on baseball's evolution. This not only enhances our comprehension of baseball, showing more robust identification of eras than past univariate time series work, but also showcases the broader applicability of multivariate change point analysis in the domain of sports research and beyond.