Using network metrics to investigate football team players' connections: A pilot study

Abstract

The aim of this pilot study was propose a set of network methods to measure the specific properties of football teams. These metrics were organized on "meso" and "micro" analysis levels. Five official matches of the same team on the First Portuguese Football League were analyzed. An overall of 577 offensive plays were analyzed from the five matches. From the adjacency matrices developed per each offensive play it were computed the scaled connectivity, the clustering coefficient and the centroid significance and centroid conformity. Results showed that the highest values of scaled connectivity were found in lateral defenders and central and midfielder players and the lowest values were found in the striker and goalkeeper. The highest values of clustering coefficient were generally found in midfielders and forwards. In addition, the centroid results showed that lateral and central defenders tend to be the centroid players in the attacking process. In sum, this study showed that network metrics can be a powerful tool to help coaches to understanding the specific team's properties, thus supporting decision-making and improving sports training based on match analysis.

Spatio-Temporal Analysis of Team Sports

“…In the sport science literature both the global and the local clustering coefficients have been applied. The idea of studying the global cluster coefficient of the players in a team is that it reflects the cooperation between players, that is, the higher coefficient for a player the higher is his cooperation with the other members of the team [21,28,65]. Fewell et al [28] also argued that a high global clustering coefficient indicates that attacking decisions are taken by several players, and thus increases the number of possible attacking paths that have to be assessed by the defence.…”

Network analysis in small-sided and conditioned soccer games

“…Utilization of technical-tactical actions, such as the pass, as a reference to the interaction between two players, has allowed patterns of cooperation to be detected at a macro level (structure of the network) and also the identification of the prominence levels of specific players during the cooperation process (micro level of analysis) (Clemente, Martins, Kalamaras, Wong, & Mendes, 2015a;Gama, Vaz, Davids, Santos, Figueiredo, & Dias, 2014;Peña & Touchette, 2012). Previous studies with official matches suggested that greater values of network density and homogeneity in cooperation were associated with the best performances (Clemente, Couceiro, Martins, & Mendes, 2014;Duch, Waitzman, & Amaral, 2010;Grund, 2012). Moreover, it has also been found that midfielders and external defenders are prominent players during the attack building and during the attack ball circulation, although greater values of degree centrality have been found for forwards in specific teams that act in counter-attack (Clemente, et al, 2015a).…”

Caracterização dos passes realizados entre jogadores brasileiros no FIFA World Cup 2014

“…Para gerar uma qualquer matriz de adjacência é necessário definir um critério de conexão entre vértices 13 . No caso do presente estudo os passes concretizados com eficácia entre companheiros foram definidos como o critério de relacionamento entre vértices seguindo estudos anteriores 5 .…”

“…Globalmente, no que se refere a métodos computacionais aplicados à análise de jogo, poder-se-á agrupar em 4 : i) análise individual cinemática; ii) análise das relações espáciotemporais; iii) métricas táticas de identificação geométrica de comportamentos organizacionais coletivos; iv) métodos baseados em redes neuronais; e v) métodos sociais baseados na teoria de grafos. No caso desta última alternativa, (comummente designado de Social Network Analysis) os métodos, apesar de computacionais pelos algoritmos utilizados, são de relativa facilidade de aplicação pela possibilidade de serem semi-computorizadas numa primeira fase de recolha de dado 5 . De facto, a possibilidade de registar notacionalmente a sequência de interações (no caso analisados regularmente pelos passes entre companheiros) e construir subsequentemente a matriz de adjacência (que resulta na rede de cooperações entre companheiros), traduz-se na facilidade com que este método pode ser generalizado para a comunidade científica e prática 6 .…”

“…De facto, um dos primeiros estudos com maior impacto na comunidade internacional foi desenvolvido no contexto do FIFA World Cup 2010, onde se identificaram os jogadores preponderantes na ação coletiva da equipa espanhola 7 . Mais tarde, a análise deixou de ser basear exclusivamente no perfil individual de proeminência na equipa, para se basear nas características gerais de interação das equipas 5,8 . De facto, em ambos os casos o contributo para o conhecimento científico e para a prática é relativamente elevada, visto as diferentes métricas existentes de social network analysis permitirem identificar padrões de interação e estilos de jogo das equipas 9 .…”