Post-Shot Expected Goals
Post-shot expected goals values a shot after it has been struck, taking placement and power into account, so only shots on target carry any value.
Why it matters. It isolates striking quality. A player whose post-shot figure beats their expected goals is placing shots well, not merely finding good positions.
What changes once the ball is struck
Ordinary xG is a before-the-fact number. It looks at where a shot was taken from, the angle available, the pressure the shooter was under. Of every shot like this, it asks, how many go in? It has no idea where the ball actually travelled, because the shot has not been struck yet.
PSxG asks a narrower, after-the-fact question. Once the shot is on its way, it looks at:
- Placement. How close to the corners, how far from the goalkeeper.
- Power. How much time the goalkeeper actually had to react.
- Trajectory. The flight of the ball rather than the shooter's position.
Two consequences follow directly from that.
- Off target shots score zero. A shot that misses the frame was never going to trouble the goalkeeper, whatever the underlying chance was worth before it was struck.
- The same chance can score very differently on each metric. A tight angle chance placed perfectly in the top corner can carry a higher PSxG than its pre-shot xG. A wide open chance blazed over the bar scores well on xG and zero on PSxG.
PSxG rewrites three verdicts
The clinical finisher. A striker whose PSxG consistently sits above their xG is placing shots into the corners, not straight at the goalkeeper. That is a striking skill, separate from getting into good positions. The gap between the two numbers is exactly where it shows up.
The great save that was not actually great. A commentator calls a stop "world class." Check the shot's PSxG. At around 0.85, the goalkeeper was heavily favoured to make that save before it even reached them, placement and power included. Reflexes still mattered, but the outcome was closer to expected than the reaction suggested.
The keeper who keeps clean sheets they should not. Add up the PSxG of every shot a goalkeeper faces across a season, then compare it with goals actually conceded. A keeper conceding meaningfully fewer than that total is outperforming what an average goalkeeper would do against the same shots. That is over and above whatever their defence achieved in front of them.
The misreadings, corrected
"PSxG is a goalkeeping stat." It measures the shot, not either player alone. Compared with goals conceded, it evaluates a keeper. Compared with pre-shot xG, it evaluates a shooter. Same number, two different jobs, depending on which side of the shot you look from.
"A big save always means the shot should have been a goal." Not necessarily. A shot with a low PSxG that gets saved was already unlikely to beat the goalkeeper. The save can look spectacular on screen while the underlying number says it was the expected outcome most of the time.
"PSxG and xG measure the same skill." They measure two skills either side of the same shot. xG measures the ability to get into a position to shoot. PSxG measures what happens once the shot is away. Confusing the two blurs shot selection with shot execution, which are coached separately for a reason.
Common questions
- Why are off target shots worth zero?
- Because post-shot expected goals asks how likely a shot was to beat the goalkeeper once it was struck. A shot that misses the target was never going to be saved, so it carries no post-shot value at all, however good the chance was.
- What does it say about goalkeepers?
- Compared against goals actually conceded, it is the standard measure of shot stopping. A keeper who concedes fewer goals than the post-shot value of the shots they faced is outperforming what a typical keeper would be expected to do.
- How is PSxG different from ordinary xG?
- xG values the chance before the shot: the position, the angle, the pressure. PSxG re-values it after the shot, using where it actually travelled. A tap-in scuffed wide keeps its pre-shot xG for that scenario, but scores zero on PSxG, because it never threatened the goal once struck.
- Can PSxG be higher than xG on the same shot?
- Yes, and that gap is the whole point of the metric. A low-probability chance from a tight angle, placed perfectly into the far corner, can carry a higher post-shot value than its pre-shot value suggested. That gap is credited entirely to execution.
- Does a header work the same way?
- The mechanics are identical: only headers that stay on target get a post-shot value. Power and placement still matter, though headers on average carry less pace than shots struck with the foot, which shows up in lower post-shot values from similar positions.
Related terms
See PSxG across the Premier League
Every player, team and match measured on the chances created rather than the scoreline.
Last updated