The Relationship Between Expected Goals (xG) Based Performance Indicators and Match Outcomes in Elite Football: The 2024–2025 English Premier League Season

Aim: This study investigates whether expected goals (xG)–based performance indicators differ systematically across match outcomes in elite football. Method: Match-level data were analyzed using 228 team–match observations (114 fixtures × 2 teams) from league fixtures involving the six teams that finished in the top six positions of the 2024–2025 English Premier League. Performance indicators included expected goals (xG), expected goals against (xGA), expected goals difference (xGD), non-penalty expected goals (npxG), finishing efficiency (goals/xG), total shots, key passes, xG per shot, and non-penalty xG per shot. Match outcome was coded as win (n = 126), draw (n = 54), or loss (n = 48). Assumption checks (normality diagnostics and Levene’s test) were conducted; group differences were examined using one-way ANOVA, followed by Tukey HSD for variables meeting homogeneity and Games–Howell for variables with unequal variances. Effect sizes (partial η²) were reported. Results: Significant group differences were found across all performance variables (all p < .001). Net and process-oriented metrics showed particularly strong discrimination across outcomes, including xGD (partial η² = .197), finishing efficiency (partial η² = .189), and npxG (partial η² = .147). Post-hoc comparisons indicated that wins were consistently associated with stronger offensive production, higher shot quality, and better conversion efficiency, while losses displayed weaker profiles on both attacking and defensive dimensions. Conclusion: Expected goals–based indicators meaningfully differentiate match outcomes among elite teams, supporting the value of process-based metrics for interpreting performance beyond final scorelines.