Betting glossary · E
Expected goals (xG) explained
Expected goals (xG) – Expected goals (xG) measure how many goals a team should have scored on average, based on the quality of its shots.
Goals in football are rare and shaped by chance. A deflected shot, a strike off the post, a great save from the goalkeeper – and the result looks completely different. Expected goals try to strip out this element of chance. Instead of counting goals, they rate every chance.
How xG is calculated
Every shot is given a value between 0 and 1. It indicates how often a shot from that situation leads to a goal on average. The basis is hundreds of thousands of shots from many leagues. Key factors are the distance from goal, the angle, the body part (foot or head), the type of assist and whether it was a counter-attack or a set piece.
Depending on the model, a penalty is worth around 0.76 to 0.79 xG. A shot from 25 metres often yields only 0.03 to 0.05 xG. A big chance from six metres, unmarked in front of goal, can be worth 0.5 xG or more. The sum of all shots gives a team's xG for the match.
What xG is useful for in betting
xG is a good gauge of a team's actual strength, better than the table after just a few matchdays. Using expected goals for and against, you can build predictions, for example via the Poisson distribution for correct scores, over/under or both teams to score.
Limits
- Different providers use different models. The figures can differ by 10 to 20 per cent.
- xG only rates shots. Dangerous situations without a shot are missing.
- The score matters: a team that leads early sits back and allows more chances – its xG record then looks worse than the team really is.
- Bookmakers know xG just as well. The fact that a team is "underperforming its xG" is usually long since priced in in the major leagues.
Related terms
- Poisson distributionThe Poisson distribution describes how likely 0, 1, 2 or more goals are when you know a team's average expected number of goals.
- Over/underWith over/under you bet on whether more or fewer goals will be scored in a match than a set line, for example 2.5.
- Both teams to score (BTTS)In the "both teams to score" (BTTS) market, "Yes" wins if each team scores at least one goal – regardless of how the match ends.
- Converting probability and oddsOdds and probability are two sides of the same number – probability = 1 / odds, fair odds = 1 / probability.
- Value betA value bet is a bet whose odds are higher than the true probability of the event justifies – it has a positive expected value.
← All terms in the betting glossary
Betting is gambling and can be addictive. 18+. Help: gamblingtherapy.org.