This post contains affiliate links. We may earn a commission at no extra cost to you.
Most casual bettors focus on the scoreline. Smart ones focus on how likely that scoreline actually was. That's where xG comes in.
If you've scrolled through football Twitter or dipped into Understat, you've probably seen "xG: 2.4" next to a match report. But what does that number actually mean, and more importantly, how can it help you find better odds?
Here's the thing: bookmakers price matches based on popularity and money flow just as much as they do on hard stats. When you understand xG and what it reveals about underlying match quality, you'll spot mismatches that most punters miss. That's value. That's where you win long-term.
In this guide you'll learn:
- Exactly what xG is and how it's calculated
- Why xG matters more than you think for betting
- How to spot xG-based value in odds before kick-off
What is xG and How Does It Actually Work?
xG stands for "expected goals". It's a metric that measures the quality of chances created in a match, not the number of chances or the final score.
Think of it this way: a tap-in from two yards out gets a high xG value (maybe 0.8). A long-range effort from 30 yards gets a low one (maybe 0.02). Every shot in a match gets assigned a probability based on hundreds of historical data points — angle, distance, defensive pressure, whether it's a header or a foot strike.
Add up all those individual shot probabilities for one team, and you've got their xG for the match.
Why xG Isn't About Luck — It's About Process
Here's where most punters get confused. A team can finish 1-0 but have an xG of 0.4. Did they get lucky? Sort of. But that's not the full picture.
What xG tells you is that they created very poor chances — low probability efforts. Over 10 matches, that team will probably regress. They won't keep winning 1-0 if they're only creating xG of 0.4 per game. The underlying performance was poor, even though the result was good.
Conversely, a team might lose 0-1 with xG of 2.1. They created quality chances. They were unlucky. Those results tend to reverse.
This is where bookmakers sometimes get it wrong. They weight recent results heavily. You weight underlying performance. That's the gap where value lives.
Real Example: Why xG Matters for Your Acca
Picture this from a typical Premier League Saturday: Arsenal play at home to a mid-table side. Arsenal have been winning 2-1, 3-2, but their xG per match has been falling. Last three home games: 2.3, 1.9, 1.5. They're winning by habit, not by dominance.
Bookmakers still price them at 1.85 to win because they've won their last three. But the stats suggest a correction is coming.
Meanwhile, the visiting side has an xG of 1.2 per game but keeps losing 0-1 or 0-2. They're creating decent chances but not finishing. You fancy them at 4.2. That's value according to xG.
Plug this into a Saturday acca and you're playing the odds that bookmakers haven't fully adjusted to underlying quality yet. That's how xG helps you find better prices than the crowd.
How Winotips Uses xG in Its AI Model
At Winotips, xG data is one of several inputs into our AI prediction engine. We don't just look at xG in isolation — that'd be lazy analysis.
Our model uses the Dixon-Coles framework, which is a Bayesian approach to rating team strength. We layer in xG, historical head-to-head records, home advantage, fixture congestion, and current form. Then we run 10,000 Monte Carlo simulations for each match. Each simulation generates a possible outcome — a scoreline, a winner, whether BTTS happens.
From those 10,000 runs, we calculate the probability of every outcome: home win, draw, away win, BTTS, over/under. We compare our probabilities to bookmaker odds and flag where the edge exists.
xG is critical because it smooths out short-term noise. A team's last three results might look terrible, but if their xG is solid, our model knows to expect improvement. That helps us spot value in underdogs before the market catches up.
Check today's picks on Winotips and compare odds at BestOdds to see how this plays out in real matches.
How to Use xG in Your Betting Strategy
Right, so you understand what xG is. How do you actually use it when you're building an acca or picking a midweek fixture?
- Check both teams' xG, not just the result. Before you fancy a team at the odds, look at their last five matches. What's their average xG created versus xG conceded? If it's much better than their results suggest, they're due a correction upwards. If it's worse, the opposite applies.
- Compare xG to bookmaker odds on the match winner. If Arsenal's xG suggests a 58% chance of winning but they're priced at 1.80 (55%), there's marginal value. If they're 1.70 (59%), you're getting worse than fair odds. Pass.
- Use xG to spot BTTS value. A match where both teams have high xG — say, 1.8 each — is more likely to see both teams score. You don't need a flashy scoreline like 2-2. A 1-1 or 2-0 both come from high-quality football. BTTS at 1.9 becomes more attractive when xG suggests both teams are clinical.
- Watch for regression on cup ties. Cup football is lower volume, so xG data is thinner. But if a lower-league side creates 0.9 xG against a Premier League club and loses 0-2, they're underrated for the replay. Use xG as a corrective lens.
- Don't ignore the defensive side. A team with high xG created is exciting. A team with low xG conceded is stability. Best value often comes from teams strong on both — they create quality and defend well. Your acca will thank you.
Frequently Asked Questions
Is xG the same as shots on target?
No. Shots on target is just a count — how many shots hit the keeper or the goal frame. xG weights each shot by quality. Three shots on target could be xG of 0.1 if they're all long-range efforts, or xG of 2.2 if they're all close-range. xG is the more useful metric for punters.
Can you make money just by betting on high xG teams?
Our model can help identify value, but no model guarantees results. xG is one signal, not the whole picture. Over-relying on any single metric will cost you in the long run. Think of xG as part of a toolkit. Combine it with form, head-to-heads, and injury news.
How do I find xG data for matches?
Understat and Wyscout are the two biggest sources. Both require subscriptions for full access, but Understat's free tier is decent for beginners. StatsBomb and Fbref also publish xG. Most sports news sites now include xG in their match reports too — scroll down on the BBC Sport or Sky Sports pages after a match.
Does xG work better for some leagues than others?
Our model can adapt, though xG is most reliable in high-volume leagues — Premier League, Championship, La Liga, Serie A. The more matches played, the more historical data underpins the xG model. Lower divisions have noisier xG data because there are fewer comparable shots to learn from. Adjust your confidence accordingly.
What xG difference should I look for to find value?
If a team's xG performance differs from their points tally by 0.5 or more per match over five games, you're looking at potential correction. A team with xG of 1.6 but results suggesting 1.1 is underperforming and likely to improve. Use this to spot good odds before they adjust.
18+ | Please gamble responsibly. Betting should be entertaining, not a way to make money. Free help: BeGambleAware.org | GamStop.co.uk | GamblingTherapy.org
Winotips provides predictions for informational purposes only. We do not guarantee any results. Always bet within your means.