Differentiate the Game Maker in Any Soccer Match Based on Social Network Approach

Abstract

A game maker controls the game by passing the ball among the teammates and is considered the most influential player in any ball-passing multiplayer team game. In the current approach, we have represented the players and the ball passing among themselves using a dynamic social network structure. Now, the challenging task is to differentiate the game maker from any match where the performance of any individual player is not directly involved with the cumulative effect of the scoring mechanism. Different temporal graph centrality metrics, such as degree, closeness, betweenness, and clustering coefficient, have differentiated the game maker in the ball-passing network. We have analyzed some random games of FIFA men’s world cup 2018, to distinguish the game makers from the matches as a case study. Another challenging problem is to locate the tactical position of the influential passes. We have investigated the games and identified those areas in different time frames. The accuracy of our method is tested over the real-life benchmark data set. The analytical results exhibit notable improvement over the existing methods. Our study opens up a new framework for the strategic analysis not only for soccer but also for similar ball-passing multiplayer team games, such as basketball and hockey.