Analyzing the dynamics of goal-scoring opportunities across football game states

In football, the flow and outcome of a match are heavily shaped by evolving game states – dynamic conditions defined by factors such as scoreline, momentum, and key events like goals and penalties. While prior studies have examined situational factors in other sports or isolated contexts in football, there is a lack of comprehensive, large-scale empirical analysis on how these game states systematically influence chance creation in football. This paper addresses this gap by investigating the impact of game states on teams’ ability to generate goal-scoring opportunities, using expected goal (xG) as a core metric across over 2000 matches from the 2023/24 season. We evaluate the role of goal difference, current match leader, and penalties, employing robust statistical tests to analyze distributions and averages under different contexts, while also examining home vs. away status, competition format, and inter-league differences. Our findings reveal that home advantage, current game state, goal difference, and penalties significantly influence attacking output, whereas competition format and league differences do not lead to substantial variations in chance creation. This paper contributes to situational modeling in football analytics by offering a large-scale, data-driven analysis of how game states affect xG generation, providing practical insights for coaches and analysts to support context-aware performance evaluation.