Flow motifs in soccer: What can passing behavior tell us?

Abstract

Passing is an important and crucial aspect of winning a soccer match. It plays a big role in important decisions made by managers and owners when buying/selling players, picking offensive/defensive strategies and defining a style of play. In this article we show how to support these decisions by analyzing the unique passing behaviors (motifs) of players and teams from the patterns in their passing-possession data. We analyze individual players as well as teams based on the diversity and frequency of their involvement in different motifs. We gather passing and possession data from 4 seasons (2012-15) of 6 big European leagues with 8219 matches, 3532 unique players and 155 unique teams and apply the network motif concept to study the patterns. By introducing an expected goals model we build on the motif concept to measure the effectiveness of styles of play. We also make use of a novel way to represent the motif data (the radar graph) to make comparisons between players and teams across multiple seasons. We show how this analysis can support scouting for players and managers, identifying unique players/teams, finding relationships between position and style and in finding a suitable replacement for La Computadora.

Evaluating the effectiveness of different network flow motifs in association football

“…Bekkers and Dabadghao (2019) investigated teams' playing styles by studying possession motifs and motifs resulting in a shot immediately after the last pass. The results are not directly comparable, as Bekkers and Dabadghao (2019) looked at the probability of scoring given that a shot is taken after a sequence of up to three passes. Their results indicate that shots taken following motifs with fewer players involved are of worse quality.…”

“…Cluster analyses and similarity measures were used to identify which players have the most similar playing styles. Bekkers and Dabadghao (2019) used network motifs to study teams' playing styles in both regular play and attacking phases of the game by distinguishing between possession flow motifs and flow motifs leading to goalscoring opportunities. Machine learning techniques were used to identify unique players, while radar graphs made comparisons between players' and teams' performance in the motifs available.…”

SoccerMix: Representing Soccer Actions with Mixture Models

“…The SPADL representation [6] addresses these concerns by converting event streams to a uniform representation designed to facilitate analysis. 2 Hence, we first transform our data into this format. Typically, playing style analysis focuses on action types and locations.…”

“…Third, ideally we would like to group actions on additional attributes such as ball direction, but considering more attributes makes each action more unique, which increases the sparsity in the data. Hence, most approaches only include one or two attributes in their analysis [2,7,14]. Rarely do approaches consider three or more attributes [8,17].…”

Guest editorial: special issue on machine learning for soccer

“…Researchers have also analyzed tactics from event stream data to find commonly occuring sequences of events that lead to attempts on goal (Van Haaren et al 2015) or identify whether an attempt is likely in the near future (Decroos et al 2017). Substantial attention has been devoted to understanding passing behaviors, particularly in terms of finding different types of recurrent passing patterns (Gyarmati et al 2014;Gyarmati and Anguera 2015;Bekkers and Dabadghao 2017). Other tasks include predicting if a pass will succeed (Spearman et al 2017), classifying different types of passes (Chawla et al 2017), and predicting whom a player may pass to (Vercruyssen et al 2016).…”