Using collective Metrics to inspect Spatio-Temporal relationships between football players

Abstract

The aim of this study was to analyse the influence of score status on the spatio-temporal relationships between team mates. Four collective metrics (weighted Centroid, weighted Stretch Index, Surface Area and Effective Area of Play) were computed based on the location of the players at each second of a match. Three matches of a team were analysed and 9218 position instants of 22 players and the ball were collected. Statistically significant differences with small effects were discovered in three possible scores in all dependent variables: Weighted Centroid y (F_{(2, 9215)}=236.627; p˂0.001; η²=0.049; Power=1.00); Weighted Centroid x (F_{(2, 9215)}=126.985; p˂0.001; η²=0.027; Power=1.00); weighted stretch index (F_{(2, 9215)}=190.005; p˂0.001; η²=0.040; Power=1.00); Surface Area (F_{(2, 9215)}=322.809; p˂0.001; η²=0.065; Power=1.00); and Effective Area of Play (F_{(2, 9215)}=139.352; p˂0.001; η²=0.029; Power=1.00). The present study showed that score status influenced the collective organisation. This is in line with previous findings which, after performing notational analysis, suggest that a team’s strategies are also influenced by the score status.

Key words: Collective behaviour; Match analysis; Metrics; Football Tactics.