Proceedings of the 5 th ESEA Conference
Edited By Oliver Budzinski and Arne Feddersen
In Urgent Need of Change? Within-season Coach Dismissals, Regression-to-the-mean, and Performance of Football Teams
Based on a comprehensive German Soccer dataset, we estimate parametric econometric models of the performance consequences of coach substitutions. In doing so, we pay special attention to the inherent regression-to-the-mean problem and present a methodological suggestion to cope with this issue in match level based analyses. Moreover, we consider two dynamic conceptions from organization theory as well as the famous “scapegoating thesis”. We identify a significant long term restoration of home advantage under the new coaches. Thus, the traditional “ritual scapegoating” interpretation of coach dismissals fails for our data. These findings are outstanding as they do contradict widely held beliefs.
football, executive succession, parametric analysis, regression-to-the-mean
The situation is well known. If an organizational unit (e.g. work team, department, division, or the whole firm) does not perform as aspired, the board or other responsible superiors are inclined to dismiss the respective executive and replace him or her with someone else. After all, the executive is responsible for the performance of the organizational unit. For boards, the dismissal of an executive is an effective way to show that actions have been taken and the problem of under-performance has been addressed. In fact, the literature maintains that poor performance is a major reason for executive succession (Kesner & Sebora 1994; Giambatista et al. 2005).
The pattern of failure induced executive change is especially visible in the team sports area where the position of the current coach is almost always put into question if...
You are not authenticated to view the full text of this chapter or article.
This site requires a subscription or purchase to access the full text of books or journals.
Do you have any questions? Contact us.Or login to access all content.