Post-Shot xG (xGOT): What It Is and Why It Matters

Expected goals (xG) is calculated at the moment a shot is taken, using inputs like distance, angle, body part, assist type, and game state. It measures the quality of the chance — how likely an average player would be to score from that position.

Post-shot xG, also called xG on target (xGOT), is calculated at the moment the ball leaves the foot or head and is heading on target. It uses the trajectory of the shot — its placement in the goal frame, its speed, its swerve — as additional inputs. xGOT measures the quality of the on-target shot, not the chance itself.

The implication: a player who consistently produces xGOT > xG is finishing well above what an average shooter would do. A goalkeeper who concedes goals worth less xGOT than the goals scored is performing below average.

xGOT models are built on shot-event data with the addition of end-location coordinates. The two best-known providers are StatsBomb (with the StatsBomb 360 frame) and Opta (with on-ball event data).

Inputs typically include the same pre-shot variables — distance, angle, body part, big chance flag — plus the placement within the goal frame (corners are higher xGOT than centre), shot velocity (where available), and on-ball context like whether the goalkeeper had been screened.

The output is a probability between 0 and 1. A close-range shot blasted into the top corner with the goalkeeper unsighted might have a pre-shot xG of 0.30 and a post-shot xGOT of 0.85 — signalling an excellent strike on a moderate chance.

A striker overperforming xGOT is hitting their on-target shots into harder-to-save areas than the model expects. Sustained over a season, this is a strong signal of finishing skill.

Underperformance can mean two things. Either the striker is hitting the goalkeeper too often (low xGOT relative to xG), or they are striking the ball well but the goalkeepers are saving it (xGOT-goals gap). The latter is variance and tends to regress; the former is technique.

• Across full seasons, top European strikers settle around xGOT ≈ 1.0 × xG with the best finishers consistently 5–15% above
• A xGOT-Goals gap > 3 over a season for a high-volume striker suggests either bad luck or strong opposition goalkeepers
• Useful for fantasy football: target high-xGOT strikers whose goal returns are lagging

Pre-shot xG tells you nothing about goalkeeper performance — it is identical regardless of who is in goal. Post-shot xGOT, by contrast, measures the difficulty of the actual on-target shot. The "Goals Prevented" stat is xGOT minus goals conceded.

A goalkeeper with positive Goals Prevented is saving more than the average keeper would have. Negative Goals Prevented over a long sample suggests genuine underperformance.

The metric is the foundation of the FIFA Best Goalkeeper award conversation and a key input in scouting models for replacing a goalkeeper.

xGOT models are still estimates. They do not perfectly capture shot velocity (most public data lacks ball-tracking), they cannot fully account for goalkeeper position, and they smooth over deflections.

Use xGOT for trend analysis over multi-match samples, not single-match verdicts. A 0.4 xGOT chance saved by an outstanding goalkeeper still counts the same as the same chance fluffed straight at them.