What is xG? A complete guide to expected goals in football
SportPulses Desk··8 min
Expected goals (xG) is the most widely used statistic in modern football analysis. What it measures, how it is calculated, how to read it in a match report, and where it falls short.
If you have read a match report or watched a TV broadcast in recent years, you have almost certainly seen “xG”. Expected goals has moved from analytics blogs into mainstream coverage because it answers a simple question better than shots or possession: how good were the chances a team actually created?
What xG measures
Every shot is given a value between 0 and 1, representing the probability that a shot of that type is scored. A penalty is usually valued at around 0.76, because roughly three in four penalties are converted. A header from a corner near the edge of the box might be worth 0.03. Add up the values of all a team’s shots and you get the team’s xG for the match: the number of goals an average finisher would be expected to score from those chances.
How it is calculated
xG models are built from hundreds of thousands of past shots. For each one, the model records factors such as distance from goal, the angle to the posts, the body part used, whether the shot followed a cross, a through ball or a set piece, and in more advanced models the positions of defenders and the goalkeeper. It then learns how often shots with those characteristics have been scored. Different data providers use different models, so the same shot can have a slightly different xG on different sites.
How to read xG in a match report
Suppose a game finishes 1-0, but the losing side had 2.1 xG and the winners 0.6. That tells you the losing team created much better chances and were either unlucky, wasteful, or up against an excellent goalkeeper. Over a single match, such gaps are common. Over a whole season, teams’ goals and xG tend to move closer together, which is why analysts use xG to judge whether a team’s results are sustainable.
Related metrics
xGA (expected goals against) applies the same idea to the chances a team concedes. npxG removes penalties, which can distort a player’s numbers. xA (expected assists) credits the player who made the pass for the xG of the resulting shot. Post-shot xG (xGOT or PSxG) also accounts for where the shot was placed, which makes it useful for judging goalkeepers: a keeper who concedes fewer goals than the post-shot xG he faced is saving more than expected.
Its limits
xG measures chances, not everything that happens in a match. It does not value a dangerous cross that nobody reaches, or a counter-attack that ends without a shot. It treats an average finisher as the benchmark, so elite strikers can outperform it for years. And because models differ, it is best used to compare teams and trends rather than to treat a single number as exact.
Why it matters to fans
Used well, xG helps explain why a team on a winning run may be heading for a dip, or why a struggling side is closer to turning a corner than the table suggests. It is not a replacement for watching matches, but it is a useful second opinion. SportPulses will use xG alongside traditional statistics in its tactical analysis.