Performance analysis tool for network analysis on team sports: a case study of FIFA Soccer World Cup 2014

• Autor(a) principal: Manuel Clemente, Filipe
• Data de Publicação: 2016
• Outros Autores: G. M. Silva, Frutuoso, M. L. Martins, Fernando, Kalamaras, Dimitris, Mendes, Rui
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.26/46736
• topic: Match analysis; software; graphical application; graphical user interface; German national team

The study of teammates’ interaction on team sports has been growing in the last few years. Nevertheless, no specific software has been developed so far to do this in a user-friendly manner. Therefore, the aim of this study was to introduce a software called the Performance Analysis Tool that allows the user to quickly record the teammates’ interaction and automatically generate the outputs in adjacency matrices that can then be imported by social network analysis software such as SocNetV. Moreover, it was also the aim of this study to process the data in a real-life scenario, thus the seven matches of the German national soccer team in the FIFA World Cup 2014 were used to test the software and then compute the network metrics. A dataset of 3032 passes between teammates in seven soccer matches was generated with the Performance Analysis Tool software, which permitted a study of the network structure. The analysis of variance of centrality metrics between different tactical positions was made. The two-way multivariate analysis of variance revealed that the strategic position (γ=1.305 ; F = 24.394; p = 0.001; η2p=0.652 ; large effect size) had significant main effects on the centrality measures. No statistical differences were found in the phase of competition (γ=0.003 ; F = 0.097; p = 0.907; η2p=0.003 ; very small effect size). The network approach revealed that the German national soccer team based their attacking process on positional attacks and not in counter-attack, and the midfielders were the prominent players followed by the central defenders. The Performance Analysis Tool software allowed the user to quickly identify the teammates’ interactions and extract the network data for process and analysis.