Moments of Competitive Balance in Major League Soccer

Abstract

In this paper, we investigate competitive balance in Major League Soccer and compare balance across talent acquisition policy regimes with a bias-corrected measure from 1996 to 2019. We evaluate multiple moments of the league talent distribution, and we add to past work by using multiple levels of aggregation that reveal heterogeneous results with respect to the distribution of talent within and across seasons. We show that there has been little improvement in competitive balance over the league's history, though there has been an increase in year-to-year stability in balance and a lower propensity for teams at the extremes of performance.

Dispersion and persistence in the competitive balance of North American major leagues 1960–2019

“…The results reported here for the NAMLs along with the previous results of Maxcy (2002) and Maxcy and Mondello (2006), as well as the results of Gerrard and Kringstad (2022) for European soccer, cast doubt on the general presumption of a strongly positive dispersion-persistence relationship. They are also consistent with the findings of Sung et al . (2022) for Major League Soccer who also conclude that their “divergent results highlight an important conceptual difference across levels of aggregation” so that “the time horizon over which we consider the distribution of talent should be central to the CB discussion in the sports economics literature going forward.” (pp.…”

“…However, there are alternative approaches to adjust outcome variances for differences in league structure. Lenten (2015) adjusts the teams' actual win percentages for the effects of an unbalanced schedule in the NFL, while Lee et al (2019b) modify the Bias-Corrected Standard Deviation measure from Lee et al (2019a) to be applicable to unbalanced schedule leagues (see also Sung et al, 2022). Lopez et al (2018) estimate the dispersion of team strength and measure the "relative randomness in game outcome" (p. 2483) among the NAMLs.…”

“…(2019b) modify the Bias-Corrected Standard Deviation measure from Lee et al. (2019a) to be applicable to unbalanced schedule leagues (see also Sung et al. , 2022).…”