The Price of Victory: A Machine Learning Analysis of How Salary Dispersion Influences Team Success in Major League Baseball

This paper examines the effect of salary dispersion on team success in Major League Baseball (MLB). Using data from 29 MLB teams over eight seasons (2012-2019), this study uses multiple models, such as Ordinary Least Squares (OLS), panel data regressions, and Random Forest machine learning, to investigate the degree – if any – that inequalities in a team's payroll impacts success, measured by winning percentage. The study's results show that, contrary to prevailing theories, salary dispersion – measured by a Gini coefficient – is not a significant predictor of a team's winning percentage (team success). Instead, team performance metrics, such as Earned Run Average (ERA), Runs Scored, and Fielding Percentage, have consistently showed to be statistically significant predictors of team success. As the results indicate that a payroll with wage inequality does not necessarily negatively impact team success, the findings suggest that MLB teams should prioritize constructing a roster that optimizes performance metrics rather than optimizes equitable salary distribution. By applying both traditional economic models (OLS, panel data regressions) and machine learning techniques (Random Forest), this study adds to the comprehensive list of literature covering organizational efficiency and labor economics in professional sports.