What the stats mean
Every number on this site, in plain English. For where the data comes from and what it can’t tell you, see how the numbers work.
How good a chance was, as a number. Each shot gets the probability that an average player scores it — a penalty is about 0.79, a 30-yarder maybe 0.03. Add them up and you get how many goals a team “should” have scored.
The same idea for the chances we give up. High xGA means opponents are getting good looks, whatever the scoreline says.
xG minus xGA. Over a season this tracks results better than goal difference, because it strips out hot and cold finishing runs.
Average shot quality. Higher means better chances, not just more of them — the difference between working the ball into the box and hopeful punts.
xG re-scored after the shot is struck, using where it's headed in the goal. A scuffed sitter loses value; a screamer into the top corner gains it. Comparing xGOT to xG shows finishing quality.
The xG of the shot a pass created. A square ball for a tap-in earns a big xA even if the striker blazes it over — it credits the pass, not the finish.
Every spot on the pitch has a danger level — how often having the ball there leads to a goal soon after. A pass or carry earns the difference between where it started and where it ended. It credits build-up that never becomes a shot.
Opponent passes allowed per tackle/interception/foul in their two-thirds of the pitch. Lower = pressing harder. It's a proxy for intensity, not a measure of whether the press is well-organised.
A situation where the shooter would reasonably be expected to score — one-on-ones, close-range finishes. Provider-flagged, so treat borderline cases loosely.
A pass that led directly to a shot (also called a chance created). Note: this is chance creation, not an assist — the shot doesn't have to go in.
A pass that moves the ball meaningfully closer to the opponent's goal — the passes that actually take you up the pitch, rather than sideways recycling.
Getting the ball into the attacking third, by pass or carry. A volume measure of territory: how often we get up the pitch, not what we do once there.
Passes attempted inside the attacking third. Different from entries: entries count arriving; this counts what we do after arriving.
Winning the ball back in the opponent's half. The point of a press — regains high up start attacks with the defence already disorganised.
Goals divided by shots. Swings wildly on small samples — a striker can look elite or hopeless over ten shots. Check the shot count before reading anything into it.
A stat scaled to per-90-minutes played, so starters and subs compare fairly. We disable it under manager or venue filters because minutes are only tracked per season.
Share of one-on-one contests won, ground and air combined.
Share of goal kicks played long rather than short — one of the clearest tells of a team's build-up intent.
Our own 0–100 blend of possession share and long-ball rate. It's a house index — the weights are ours, so treat it as a description, not an official stat.
Whether two things move together, from −1 to +1. It does not prove one causes the other — teams that are winning pass more; passing more doesn't make you win.
n is the sample size. “Not significant” means the sample is too small to rule out coincidence — we grey those out rather than hide them, and never colour a correlation red unless it's significant.
A goal with no assist pass to track — dribbled moves, loose balls, rebounds. In build-up charts these count as direct (0–1 passes) because there was no passing move to measure.