What Is xG (Expected Goals) in Soccer and How to Use It
Turn on a broadcast of almost any major match today and at some point a graphic will flash across the screen showing each team’s expected goals total. The stat, universally shortened to xG, has become the most widely cited advanced metric in soccer analysis, but plenty of fans who hear the term regularly still aren’t entirely sure what it measures or how it’s calculated.
The Basic Concept
Expected goals assigns a probability value, somewhere between 0 and 1, to every single shot taken in a match, representing the likelihood that an average professional player would score from that exact position and situation. A close-range tap-in from inside the six-yard box with an open goal might carry an xG value close to 0.9, meaning a player in that position would be expected to score roughly 90 percent of the time. A speculative strike from 30 yards out with a defender closing down might carry an xG value closer to 0.03, reflecting how rarely shots from that range and pressure actually go in.
Add up the xG values for every shot a team takes across a match, and you get that team’s total expected goals for the game, a number that represents roughly how many goals an average team would be expected to score given the exact same shots.
What Factors Actually Go Into the Calculation
Modern xG models are built from massive historical datasets covering hundreds of thousands of shots across professional leagues worldwide, and they weigh a range of factors for each individual attempt. Distance from goal and the angle of the shot relative to the goal frame are the two most heavily weighted variables, since both dramatically affect scoring probability on their own. Beyond that, models typically account for whether the shot was taken with a player’s stronger or weaker foot, whether it was a header, whether the shot came directly off a through ball, cross, or rebound, and how many defenders were positioned between the shooter and the goal at the moment of the strike.
Why xG Became Popular With Analysts
Before expected goals gained mainstream traction, most statistical analysis of attacking performance relied on raw shot counts or shots on target, both of which treat a tap-in from a yard out identically to a wildly speculative effort from near the halfway line. xG solved that problem by weighting shots according to actual quality, giving analysts a much better proxy for how many goals a team’s underlying performance should have produced, independent of finishing luck or a goalkeeper’s individual brilliance on a given day.
This makes xG especially useful for spotting teams that are over- or under-performing their underlying numbers. A team consistently scoring well above its cumulative xG total over many matches is often relying on finishing quality that regression suggests won’t hold up indefinitely, while a team scoring well below its xG is frequently unlucky or missing a clinical finisher, and tends to see results improve if the underlying chance creation stays consistent.
The Limitations Worth Knowing
xG is a probabilistic model, not a certainty, and it says nothing about the specific quality of the finish itself, only the situation the shot was taken from. A world-class striker curling a shot into the top corner from a tough angle carries the same xG value as an average player shanking a shot from the identical spot, because the model can’t account for individual technique once the shot is struck. It also can’t fully capture context like a goalkeeper being unusually out of position, or a defender getting a slight deflection that changes a shot’s trajectory.
Reading xG the Right Way
The most useful way to read an xG chart is as a season-long or multi-match trend rather than a single-game verdict. Any individual match can be won or lost by a moment of brilliance or a defensive lapse that no model could predict, but a team’s cumulative xG difference across ten or fifteen matches tends to correlate closely with where that team actually ends up in the table over a full season.
