Network analysis in small-sided and conditioned soccer games

Abstract

The purpose of this article was to investigate the influence of additional players and playing position on the network properties during 2x4 minutes small-sided and conditioned games (SSCG) in soccer. Eighteen young soccer players (age 16.4±0.7 years), six defenders, six midfielders, and six forwards, voluntarily participated in SSCGs with different task conditions (4vs.3, with an additional player inside the pitch, 3vs.3+2, with two support players at the side of the pitch, and 3vs.3, numerical equality). General (density, total links and clustering coefficient) and individual (degree centrality, degree prestige, and page rank) network properties were analyzed using the SocNetV® software. Results showed higher values of density (F=59.354, p=.001), total links (F=40.951, p=.001), and clustering coefficient (F=21.851, p=.001) during the 4vs.3 SSCG. Besides, midfielders showed higher values of degree centrality than defenders and forwards (F=10.669, p=.001). Midfielders and forwards also showed higher values of degree prestige than defenders (F=5.527, p=.005). These results indicate that both task condition and playing position influence the general and individual network properties during SSCGs. For this reason, it is suggested that both task condition and team composition need to be adjusted to the coaches’ purpose for each training session in order to maximize the possibilities of cooperation among the teammates.