Macrostructural Passing Networks Across Competitive Contexts: A Case Study of Manchester City

Although Social Network Analysis has become an established approach for investigating collective behaviour in football, research has predominantly focused on player-centred metrics, with comparatively limited attention given to the longitudinal evolution of macrostructural passing networks across different competitive contexts. This study investigated whether the macrostructural organisation of Manchester City’s passing network varied according to season phase, match location, quality of opposition, and match outcome throughout the 2023–2024 English Premier League season. Match event data from all 38 league matches were obtained through Wyscout®, and successful passing interactions were transformed into directed weighted adjacency matrices. Macrostructural Social Network Analysis was performed using uPATO®, computing Total Links, Network Density, Average Distance, Network Diameter, Network Heterogeneity, Transitivity, Reciprocity, Global Centralization, Global Prestige, and the Assortativity Coefficient. Descriptive and inferential statistical analyses were conducted using IBM SPSS Statistics® (Version 29.0) to compare macrostructural passing network metrics according to season phase (first vs. second half of the league), match location (home vs. away), quality of opposition, and match outcome. No statistically significant differences were observed according to season phase or match location (p > 0.05). However, statistically significant differences were observed across quality-of-opposition groups for Total Links and the Assortativity Coefficient, and across match-outcome groups for Total Links and Reciprocity. These findings indicate that contextual variation was associated with selected macrostructural network properties rather than with broad differences across the metrics examined. This case study highlights the value of examining collective passing organisation across different competitive contexts using macrostructural network metrics.