xG, short for expected goals, measures the quality of the chances a team or player creates. Every shot is given a value between 0 and 1 that reflects how often similar shots have been scored in the past. Add those values up and you get a figure for how many goals the chances were worth.
This guide explains what xG means, how it is calculated, what xGA and other related numbers show, and how to read it sensibly when you are looking at form. It sits alongside the rest of our football betting explainers.
What does xG mean in football?
xG answers a simple question: how likely was that chance to be scored? A shot with an xG of 0.1 is the kind of chance an average player would score about one time in ten. A shot with an xG of 0.5 goes in about half the time.
The idea is familiar to anyone who has heard a commentator say a striker “scores that nine times out of ten”. xG puts a number on that judgement, using data from a very large number of past shots rather than gut feeling.
A team’s xG for a match is the total of all its shots. If a side has an xG of 1.8, the chances it created would, on average, produce just under two goals. It does not mean the team should have scored exactly 1.8, because nobody scores a fraction of a goal.
How is xG calculated?
Each data provider builds its own model, trained on a big database of historical shots and whether they went in. The model looks at the circumstances of a new shot and compares it with similar ones. Opta, for example, says its model weighs more than 20 variables. Among the most important are:
- Distance to goal: the closer the shot, the higher the value, as a rule.
- Angle to goal: a shot from a tight angle is harder to score than one from straight in front.
- Shot type: a header is usually worth less than a shot with the foot from the same spot.
- Pattern of play: open play, a fast break, a corner or a direct free kick.
- The pass before the shot: a through ball and a cross create different kinds of chance.
- Defenders and goalkeeper: how much pressure the shooter is under and where the keeper is standing.
Penalties are treated separately because they are so consistent. Opta gives every penalty the same value, 0.79 xG, based on how often penalties have been converted historically.
Because the models differ, two websites can show slightly different xG figures for the same match. Stick to one source when you compare teams, so you are always measuring with the same ruler.
What is expected goals against (xGA)?
Expected goals against, or xGA, is the same measure from the other side. It totals the xG of every shot a team has conceded, so it shows the quality of chances it is allowing the opposition.
Put the two together and you get a picture of a team’s underlying performance. A side that regularly creates more xG than it concedes is usually getting the better of games, even if a run of results says otherwise. Some sites show this as xG difference, the xGA total taken away from the xG total.
Other xG numbers you might see
- npxG: non-penalty expected goals. Penalties are removed, which gives a fairer view of what a team or player creates from open play and set pieces.
- xG per game or per 90: the season total divided by matches or by 90-minute periods played, so players and teams with different minutes can be compared.
- xA or expected assists: credits the passer with the xG of the shot their pass set up. Definitions vary between providers.
- xGOT: expected goals on target. It only looks at shots that hit the target and takes account of where in the goal the ball was heading.
What does it mean when a team outperforms its xG?
If a team scores more goals than its xG, it has finished its chances better than an average side would. That can come from genuinely good finishers, but over short spells it often comes down to variance: a few deflections, a long-range strike or two, or an opposition keeper having a bad day.
The reverse is true for a side scoring well below its xG. Poor finishing, an in-form goalkeeper on the other side and simple bad luck can all play a part. Over a larger sample of matches, team scoring tends to sit closer to the underlying numbers.
One common mistake is to assume an overperforming team is now “due” a run of underperformance. Opta points out that this is the gambler’s fallacy: goals already scored are banked, and the future is not obliged to even things out.
How to use xG when you look at football form
The league table only tells you about results. Set it alongside xG and xGA and you can see which teams are playing better or worse than their position suggests. Our up-to-date football league tables are a good place to start before you compare the underlying figures.
xG is also useful context for goals markets. A fixture between two sides that both create plenty of chances is a different proposition from a game between two teams that rarely threaten. Our guide to how over and under goals markets settle explains those bets, and the guide to whole, half and quarter-goal handicap lines covers handicaps.
Keep its limits in mind, though:
- One match is a small sample. A higher xG total does not mean a team deserved to win. It measures chance quality, not the expected result.
- Game state matters. A team that goes two goals up early may sit back, so its xG drops and the opponent’s rises.
- It only counts shots. A dangerous cross that nobody gets a touch on adds nothing to xG.
- Team news changes things. xG built up with a key striker in the side tells you less once that player is injured.
Used with form, team news and a sensible view of the price, xG is a helpful extra layer. You can see how we approach the weekend’s games on our Premier League betting tips page. No statistic guarantees a result, and no approach guarantees a profit.
Frequently asked questions
What does xG mean in football?
xG stands for expected goals. It gives every shot a value between 0 and 1 based on how often similar shots have been scored, so it measures the quality of chances.
How is xG calculated?
A model compares each shot with a large database of past shots, using factors such as distance, angle, shot type, the pass before it and defensive pressure. Each provider uses its own model.
What is a good xG in a match?
There is no fixed benchmark. The useful comparison is a team’s xG against its opponent’s in that game, and against its own average over the season.
What is expected goals against (xGA)?
xGA adds up the xG of every shot a team concedes. It shows the quality of chances the team is allowing.
What does it mean when a team outperforms its xG?
It has scored more goals than its chances would usually produce. That can reflect good finishing, but over short spells luck often plays a big part.
Further reading: our explainer on how both teams to score bets settle, and Opta Analyst’s guide to expected goals, which sets out how its xG model works.
18+ only. Please gamble responsibly and read our responsible gambling guidance. For free, confidential support visit BeGambleAware.org.
Leave a Reply