Passing Decisions in Football: Introducing an Empirical Approach to Estimating the Effects of Perceptual Information and Associative Knowledge

Abstract

The importance of various information sources in decision-making in interactive team sports is debated. While some highlight the role of the perceptual information provided by the current game context, others point to the role of knowledge-based information that athletes have regarding their team environment. Recently, an integrative perspective considering the simultaneous involvement of both of these information sources in decision-making in interactive team sports has been presented. In a theoretical example concerning passing decisions, the simultaneous involvement of perceptual and knowledge-based information has been illustrated. However, no precast method of determining the contribution of these two information sources empirically has been provided. The aim of this article is to bridge this gap and present a statistical approach to estimating the effects of perceptual information and associative knowledge on passing decisions. To this end, a sample dataset of scenario-based passing decisions is analyzed. This article shows how the effects of perceivable team positionings and athletes' knowledge about their fellow team members on passing decisions can be estimated. Ways of transfering this approach to real-world situations and implications for future research using more representative designs are presented.

Using position data to estimate effects of perceptual features of play on passing decisions in soccer

“…To date, studies analyzing the effects of these various contextual features on real-world passing decisions simultaneously are yet missing. Recently, methods for examining relations between the mentioned contextual features and passing decisions have been presented (Steiner, 2018). To illustrate the methods, a sample dataset of passing decisions made in static game scenarios was used.…”

“…Obviously, the relation between contextual information and behavior decisions will remain unknown unless actual behavior is also measured (Baumeister et al, 2007). The aim of this study was thus to replicate Steiner's (2018) analysis using data of passes played during real football competitions. We expected that the same four kinds of contextual information would be similarly related to passes played during football competitions.…”

Predicting Pass Receiver in Football Using Distance Based Features

“…Previous approaches predicting the receiver of a pass use different types of features. While the use of qualitative features (relative positioning of players, with respect to the sender and the receiver of the pass) is recommended [7], methods based on quantitative features (distances between players and pass lanes) yield similar results in the prediction of the pass receiver [5].…”

“…This study is based on quantitative features, and shows pass prediction results similar to [7]. It should be noted that the data used in [5] is not based on real game data, but on scenarios that are evaluated by football players in an offline setting. This might have an influence on the pass decision process of the player.…”

“…In [5], Steiner studies the effects of perceptual information on the probability for a player to receive a pass. He concludes that players with open passing lanes have a higher chance of receiving a pass.…”