Mean reversion or a breath of fresh air? The effect of NFL coaching changes on team performance in the salary cap era

Abstract

Just as firms must consider the impact of changes in management, sports teams must consider whether a coaching change will improve the team’s on-field performance. I examine the effects of coaching changes for National Football League (NFL) teams between the 1995 and 2012 seasons. A variety of factors contribute to an NFL team’s performance reverting towards league-average levels between seasons. Thus, a nominal improvement in team performance could be due to improved management or it could simply be mean reversion. I find that, after accounting for the highly significant mean reversion effect during this time period, firing a coach reduces a team’s expected performance during the next season and the team’s average performance over the next two seasons. This effect is present when wins are used to measure performance, but also when performance is measured by point differential and playoff appearances, two variables that avoid some shortcomings of using team wins as a measure of performance. I conclude that teams are firing coaches an inefficiently high percentage of the time.

Keywords:

• managerial change
• NFL
• team performance
• mean reversion

JEL Classification:

• J63
• L83

Notes

¹  Data were taken from pro-football-reference.com