Competitive Balance in North American Sports Leagues: An Application of Multivariate Change Point Analysis

We use recent methodological developments in change point analysis that estimate structural changes in multivariate time series – both in level (mean) and variance – adding to the literature on long-term North American sports league competitive balance tracking. Our method estimates changes by simultaneously examining 9 competitive balance measures that include game uncertainty, playoff uncertainty, and consecutive season uncertainty classifications. This approach provides insight into structural changes in competitive balance not found in other work while also identifying the competitive balance measure(s) most responsible for estimated change points. Additionally, we identify changes to the variance structure of balance, which was not previously considered in the structural change literature. Overall, rival leagues, labor supply increases, expansion, and changes to playoff structures are most commonly concurrent with structural changes. We suggest relevant research extensions for sports economists.