By plotting xG for/against over a full season with vertical markers at manager change dates, you can visualize whether a new manager's style produced measurable changes in the quality of chances created and conceded — separate from result variance. This is cleaner than using results because it removes luck. A manager who improves the process (xG differential) may not show up in results immediately; conversely, a manager whose team is winning on poor xG is a regression risk.
Construction: (1) compile match-by-match xG for and against; (2) plot as a rolling average (5-match window is typical); (3) add vertical blue lines at manager appointment dates; (4) visually inspect whether xG for/against trends shifted after each appointment. For rigorous analysis, use a statistical change-point detection method (e.g., CUSUM or Bayesian change-point) rather than visual inspection. Present alongside actual results to show where process and outcome align vs. diverge.
When a manager changes, the team's xG creation and concession profiles shift measurably within 3-5 matches, not the 15-20 match "settling in" period that conventional wisdom assumes. The reason: the new manager immediately changes pressing triggers, defensive line height, and buildup routing — all of which show up in spatial xG patterns well before results stabilize. The results lag because variance is high in small samples, but the PROCESS shift is immediate.