Expected goals (xG) is a stat that measures the quality of chances a team creates, by assigning every shot a probability of being scored. For bettors, it is one of the most useful tools available: it tells you whether a result was deserved or lucky, and it helps you spot teams the market is mispricing before the odds catch up. Used properly, xG is a way to find value — not a crystal ball.
What expected goals (xG) actually measures
Every shot in a match can be given a probability of ending up in the net, based on the characteristics of that chance and the events leading up to it. xG is usually expressed as a number between 0 and 1. A tap-in from two yards might be worth 0.9 xG, while a speculative effort from 30 yards might be worth 0.03. Add up the xG of every shot a team takes and you get their xG for the match — a measure of how many goals an average team would expect to score from those chances.
The models behind xG are built on historical data from hundreds of thousands of shots. Common inputs include:
- Distance and angle to goal — the two biggest drivers of shot quality.
- Body part — headers convert at a lower rate than shots with the feet.
- Type of assist — a cut-back or through ball creates a better chance than a cross.
- Game situation — open play, set piece, fast break, or penalty.
- Defensive pressure — the number of defenders between the shooter and goal.
A useful anchor: a penalty is worth roughly 0.76 xG, because historically around 76% of penalties are scored. That single number shows the logic of the whole system — xG is just the long-run conversion rate of a given type of chance.
Why xG beats the scoreline for finding value
Final scores lie. A team can win 1-0 from a single half-chance while being battered, or lose 1-0 having created a hatful of openings. Goals are rare events, so a 90-minute scoreline is a small, noisy sample. xG smooths that noise by judging teams on the quality and volume of chances they create and concede, which tends to be far more repeatable from week to week.
That matters because bookmakers and recreational bettors both overreact to recent results. If a strong side loses two games while posting healthy xG numbers — plenty of good chances created, few conceded — the market may drift their price out further than it should. That gap between the true probability and the offered price is exactly what value betting is built on.
How to use xG to find value bets
The practical workflow is straightforward, even if the judgement behind it takes time to build:
1. Compare xG to actual goals over a run of games
Look at a team’s xG for (chances created) and xG against (chances conceded) over the last 8–12 matches, not just one. A side scoring well above its xG is likely to regress — finishing hot streaks rarely last. A side scoring below its xG may be due a correction upward. The same logic applies at the defensive end.
2. Build your own rough expectation, then check the odds
If a team consistently out-creates opponents by a clear xG margin but their match-result or over/under goals price does not reflect it, you may have an edge. xG per game also feeds naturally into goals markets and both-teams-to-score bets, because it describes how many clear chances each side tends to generate.
3. Convert your estimate into implied probability
To judge value you need to compare like with like. Turn the decimal odds into implied probability (1 ÷ odds) and compare it to your own estimate. If you believe a team is genuinely a 55% chance and the implied probability from the price is only 48%, that is a potential value bet. Understanding how prices translate into probability is covered in our guide to betting odds explained.
The limitations — be honest about them
xG is powerful, but treating it as gospel is where bettors lose money. Keep these limits front of mind:
- It ignores finishing and goalkeeping quality. xG assumes a league-average finisher. A genuinely elite striker or a world-class keeper will consistently beat the model — and that is a skill, not just luck.
- Small samples mislead. One or two matches of xG tell you almost nothing. You need a run of games before the numbers stabilise.
- Models disagree. Different providers use different inputs, so one site’s xG will not match another’s. Pick a source and stay consistent.
- Context is missing. xG does not know a team played 60 minutes with ten men, chased a game late, or rested players before a cup tie. You still have to watch football.
- The market already uses it. Sharp bookmakers price xG-style models in. The edge comes from applying context and judgement the raw number misses, not from the number alone.
Putting it together: a sensible xG routine
Treat xG as one input among several. Start with underlying xG trends to form a view, then layer in team news, motivation, schedule congestion, and tactical match-ups. Only bet when your estimate of the true probability is clearly better than the price on offer, and stake sensibly using sound bankroll management rather than piling in because one stat looks attractive. Consistency and discipline beat any single metric.
Frequently asked questions
What is a good xG number for a team in a match?
There is no universal figure, but a team posting around 1.5–2.0 xG in a game has usually created a healthy set of chances, while under 0.5 xG suggests they struggled to threaten. Judge it relative to the opponent and over several matches rather than in isolation.
Can you make money using only xG?
No stat guarantees profit. xG improves your reads and helps you spot mispriced teams, but bookmakers factor it in too. Long-term success still depends on finding genuine value, adding context the model misses, and managing your stakes.
Where can I find xG data?
Several analytics sites and broadcasters now publish xG for major leagues, and many bookmakers display it alongside stats. Choose one reputable source and use it consistently, since figures vary between models.
Is xG useful for lower leagues?
It can be, but models are trained mostly on top-division data, and coverage in smaller leagues is thinner and less reliable. Treat lower-league xG with more caution and lean harder on your own research.
The takeaway
xG is one of the best tools a modern bettor has for judging which teams are genuinely good and which have been flattered or unlucky by the scoreline. Use it to spot regression and mispriced sides, always convert your view into implied probability before betting, and never let one number override context and judgement.
Bet responsibly — 18+, only stake what you can afford to lose, and seek help via a service like GamCare or BeGambleAware if gambling stops being fun.
About the Author
