Passing Networks: An Examination of the Prevalence of Positions in Positional Attack in Handball

Abstract

This study aimed to examine the player positions that are more dominant in positional attack. A total of six international matches were selected, all matches involving MOL-Pick Szeged handball club. For the 12 graphs obtained from social network analysis of MOL-Pick Szeged and each of their six opponents, a set of network metrics were computed. Data were calculated using Gephi software. There were an average of 745 (SD = 68) passes for each game constituting a total of 4,472 passes between players with an average of 373 (SD = 78) passes coded per team. Computations involved micro level of analysis related to player position centralities across the team, and macro levels of analysis related to the network of the team as a whole. The results showed that the backcourt players are the key players in the attack against the most used defences. Interestingly, a comparison of total passes per match against 6:0 defences and against 5:1 defences, for Pick and the opponent teams, showed that the total passes for Pick were only higher than their opponents in the match that Pick lost. The total passes against 6:0 formation were only higher for Pick in one match, and the total passes against 5:1 formations were higher for Pick in the matches that the team drew and lost. These results may indicate that passing network metrics can identify weaknesses of a team to particular defensive formations used by their opponents.