Moneyball in Offensive vs Defensive Actions in Soccer

Abstract

There is an established tradition in soccer society, for both soccer fans and club managers to value forwards more than defensive players. However, the soccer rules imply an equally important role of goals scored and goals conceded in a team win. This paper employs these facts to formulate the research hypothesis of undervalued defensive, compared to offensive actions, by professional soccer clubs, known as Moneyball phenomenon in sports economics literature. To test our hypothesis, we use two separate data sets at team and player level (1,224 and 776 observations correspondingly) from two seasons (2017-2019) of the German Bundesliga. We estimate the two groups of models with a dependent variable being, correspondingly, an indicator of win and a market value. We keep the set of controls as similar as possible to make the results of the two groups of models comparable with each other, in terms of a relative contribution of offensive and defensive actions. Offensive actions are measured by shots and key passes, while tackles, interceptions and clearances stay for defensive actions variables. All the key variables are normalized, and the resulting estimates demonstrate both “absolute” and “relative” Moneyball in offense vs defense in different model specifications. This is the first introduction of a term and method of a “relative” Moneyball, to our knowledge. In addition, our results show that there is room for improvement for Bundesliga clubs’ cost of win efficiency, in redistributing funds from offense to defense, at least to some extent