This post contains affiliate links. We may earn a commission at no extra cost to you.
Most UK punters focus on who scored. Smart bettors focus on who should have scored. That's where xG comes in. Expected Goals — or xG — is the metric that separates casual betting from data-driven betting. It doesn't predict match results. It exposes whether the odds you're looking at are actually value or just noise.
If you've scrolled through a Sky Sports match report or caught a pundit mention "that was a low xG performance," you've heard this term. But what does it actually mean for your Saturday acca? Why should you care if one team mustered 1.2 xG while another hit 3.7? And how do you turn that number into better betting decisions?
Here's the truth: bookmakers price matches based on many things — team form, injuries, betting volume. What they sometimes miss is the gap between a team's luck and their actual quality. xG helps you find that gap.
In this guide you'll learn:
- Exactly what xG is and how it's calculated
- Why xG matters more than the scoreline for betting value
- How to spot overpriced teams using expected goals data
What is xG and How Does It Work?
xG stands for expected goals. Rather than counting goals scored (which is often just luck), xG measures the quality and quantity of chances a team created. Every shot a team takes gets assigned a probability — based on its distance, angle, and defensive pressure — of becoming a goal. Those probabilities add up to a team's total xG.
Think of it this way: if a team takes a shot from two yards out with no defenders in the way, that shot gets a high xG value — maybe 0.65. If they're taking a 25-yard pot shot from a tight angle, that gets a low value — maybe 0.03. Add up all the shots in a match, and you've got the team's xG total.
Over a season, xG aligns remarkably well with actual goals scored. A team that outperforms its xG by a lot in September often underperforms it by January. Football evens out. Luck doesn't last. Quality does.
Why xG Matters More Than the Scoreline
A team can win 2-1 with 0.8 xG while the loser had 2.1 xG. Who played better? The loser. Who did the bookies price correctly for the next match? Probably not the winner — because the next set of odds will assume that team is stronger than the data actually shows. That's where value hides.
Arsenal plays West Ham. Arsenal wins 1-0 but xG shows West Ham 1.8, Arsenal 0.9. The narrative says Arsenal are clinical. The data says West Ham were unlucky. When the bookies line up the next match, they might still be overpricing Arsenal because they're anchored to that 1-0 scoreline. If you're savvy enough to check the xG, you spot it first.
How xG is Calculated
Models vary slightly — StatsBomb, Understat, WhoScored all have their own versions — but the principle is the same. A shot's xG value depends on:
- Distance from goal — closer shots = higher xG
- Angle — central shots = higher xG than wide ones
- Defensive pressure — how many defenders are between the shooter and goal
- Shot type — headers, volleys, penalties all have different conversion rates
- Previous pass quality — was the ball moving fast or slow?
Professional models look at hundreds of thousands of historical shots and ask: "Of all the shots from this position, angle, and situation, what percentage went in?" That percentage becomes that shot's xG. A 0.15 xG shot goes in 15% of the time historically. A 0.75 xG shot goes in 75% of the time.
Here's a concrete example. Liverpool plays Brighton at home. Liverpool creates three chances: a 6-yard tap-in (0.72 xG), a 12-yard shot with one defender nearby (0.18 xG), and a 30-yard long-range effort (0.04 xG). That's 0.94 xG total. They score once. Underperformance. Brighton creates two chances: a 16-yard shot (0.12 xG) and a 22-yard effort (0.06 xG). That's 0.18 xG total. They score zero. Exactly as expected.
The final score is 1-0 to Liverpool. That looks decisive. The xG (0.94 vs 0.18) tells you Liverpool dominated in quality. Both numbers are useful — but only one helps you price the next match correctly.
How Winotips Uses xG in Its AI Model
At Winotips, xG isn't just a fun stat to know. It's baked into our AI predictions from the ground up. We don't just look at last season's results or recent form — we combine xG data with team strength, fixture difficulty, and player availability to build probabilistic forecasts.
Our model runs 10,000 Monte Carlo simulations for every match using the Dixon-Coles framework, a mathematical approach that weights recent performance more heavily than older data. xG feeds into that model because it's a cleaner signal of team quality than goals alone. A team that consistently outperforms their xG over several seasons? That's genuinely good finishing. A team that outperforms by 0.5 goals over two matches? That's usually luck correcting itself.
When you check today's AI predictions on Winotips, you're seeing forecasts that account for expected goals, not just past scorelines. Compare odds at BestOdds and you'll see where bookmakers are pricing teams based on recent results — often missing the xG story underneath.
The practical upshot: our model can identify matches where public perception (and therefore odds) misaligns with underlying quality. That's where consistent value lives.
How to Use xG in Your Betting
Using xG to improve your bets doesn't require a PhD in statistics. Here are the steps:
1. Check xG after every match you're considering
Before you fancy a team for Saturday's acca, pull up their last three matches and look at their xG — not just goals. If Manchester City beat Fulham 3-1 but xG showed City 1.2 and Fulham 0.6, City got lucky. Their next odds might still assume that level of dominance, even though the underlying quality was much tighter. That's your edge.
2. Look for xG overperformers and underperformers
A team that's beaten xG by 0.8+ goals over their last three games is likely to regress. Odds might still price them as if they're that strong. Conversely, a team that's underperformed xG is probably due a run of wins. Odds might underestimate them.
3. Use xG to filter your acca selections
Building a midweek cup-tie acca? Don't rely on form alone. A Championship team that's recently overperformed xG massively might look hot, but the odds won't reflect that weakness — yet. Teams with consistent xG outperformance (goal difference) are safer bets than teams with lucky recent results.
4. Compare xG against bookmaker odds
If our model suggests a team has a 65% win probability based on xG and shot-creation data, but the bookies are pricing them at 1.85 (54% implied probability), that's value. Our model can help identify those spots — compare options with BestOdds to lock in the best available price.
5. Track xG trends, not individual matches
One match's xG is noise. Three matches' xG is a trend. A team can have a 0.6 xG outlier and still be genuinely strong. But if a team averages 0.4 xG per match over five games while their goals tally looks respectable, that's a regression warning waiting to happen. Build that into your long-term selections.
Frequently Asked Questions
Is xG more accurate than goals scored?
No — xG and goals tell different stories. Goals are what happened. xG is what should have happened based on quality. Over a full season, xG predicts future performance better than goals alone. But in a single match, goals are facts. xG is useful context for understanding whether those goals were deserved or lucky. Our model uses both.
Can you make money using xG?
xG can help identify value odds — where bookmakers misprice a team because they're anchored to recent results rather than underlying quality. Our model can help spot those discrepancies, but no model guarantees profit. Football is unpredictable. xG just gives you better information than the scoreline alone.
Do all bookmakers ignore xG?
No — the biggest, most sophisticated shops factor xG into their pricing. But there's often a lag. A team's xG might shift significantly in the last 48 hours before a match, but betting odds might not move fast enough to reflect it. That's where punters with early information can find edges. Check xG updates before you lock in your acca.
What's a "good" xG number for a winning team?
There's no magic number. A team winning 1-0 with 0.3 xG might've had a world-class goalkeeper performance — that's still a good result. A team winning 3-1 with 1.1 xG overperformed massively. Context matters. Generally: if you're the underdog and your xG exceeds the favourite's, the odds underestimated you.
Should I use xG alone to decide a bet?
Never. xG is one input among many — team news, head-to-head record, fixture congestion, pitch conditions all matter. A team might have elite xG but be missing four key defenders. Our model weighs xG alongside dozens of variables. Think of xG as the foundation, not the whole building.
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.