The Number That’s Changing How People Bet on Football
Here’s something most casual bettors don’t know: a team can win a match 1-0 and still have played worse than the losing side. Sounds odd, right? But that’s exactly what xG football stats reveal. Expected goals – a metric born in sports analytics labs and now used by coaches, analysts, and sharp bettors worldwide – strips away luck and shows what the scoreline probably should have been.
And the betting implications are huge.
What Are Expected Goals, Exactly?
Expected Goals Explained in Simple Terms
An expected goal (xG) is a number between 0 and 1 assigned to every shot in a football match. It measures the probability that a specific shot results in a goal, based on factors like shot location, angle, assist type, and whether it was a header or a foot shot. A penalty kick, for example, carries an xG of around 0.76. A header from 30 yards out might get 0.02.
So when a team creates five shots with a combined xG of 2.4, it means those chances – statistically – should produce about two or three goals. If the actual score was 0-0, something unusual happened. Maybe the goalkeeper had a great game. Maybe a striker missed an open net. Either way, the xG tells a different story than the final result.
The metric was developed and popularized in football over the past decade or so, drawing on millions of historical shots to build probability models. It’s not perfect. But it’s probably the most useful single number in modern football analytics.
What xG Does NOT Measure
xG doesn’t account for everything. It can’t measure goalkeeper quality directly, or the mental state of a player in a high-pressure moment. Some models also don’t capture shot technique or the “quality” of the pass that created the chance. So there’s always a margin of error baked in.
But here’s the thing – for predicting scores across a whole season or a series of matches, xG is far more reliable than actual goals. Luck evens out. xG doesn’t lie as often.
How to Use xG Data for Score Predictions
Reading Match-Level xG Numbers
The first step is learning how to read a basic xG summary for a match. Most data sites present it as two numbers – one per team – like this:
| Team | Shots | xG | Actual Goals |
| Team A | 14 | 2.1 | 1 |
| Team B | 7 | 0.6 | 2 |
In this example, Team B won 2-1. But Team A dominated on expected goals. This kind of result – where the xG “loser” wins the actual match – happens more often than people think, somewhere around 25-30% of matches depending on the sample.
For bettors, this matters a lot. Team A in this scenario is likely underrated going into their next fixture. Team B may be overrated. And if you’re thinking about how to predict football scores more accurately, spotting these gaps between xG and actual results is a great starting point.
Looking at Season-Long xG Trends
Single matches can be noisy. Across a whole season, patterns become clearer and much more useful.
A team sitting in mid-table might actually be generating xG numbers that rival a top-four side. That’s a team probably due for better results. Conversely, a team near the top might be massively overperforming their xG – scoring more than their chances suggest – which means they’re probably riding some luck and could drop off.
Smart bettors look at “xG difference” (xG For minus xG Against) as a rough quality indicator. Teams with strong positive xG differences tend to win more games over time, even if their current league position doesn’t reflect that yet. You can find this data broken down by team on sites like Understat or FBref. Spend a few minutes with those numbers before placing a bet and you’ll already be thinking differently about a fixture.
Applying xG to Correct Score Betting
This is where things get interesting. Correct score markets are notoriously hard to win, but xG gives bettors a semi-structured way to reason about likely scorelines.
If Team A has an average xG of 1.8 per match and Team B concedes an average of 1.4 xG against, you’d expect somewhere around 1-2 goals from Team A in a home game. That’s not a guarantee – it’s a range to work with. Combine that with Team B’s attacking xG and you can build a rough probability map of possible scorelines. Something like 2-1 or 1-1 might emerge as the statistically favored outcomes, even if the bookmaker’s odds don’t fully reflect that.
BetFury is one platform where bettors can find correct score markets alongside a range of football betting options worth exploring with this kind of data-driven approach.
Common Mistakes When Using xG for Betting
Treating xG Like a Guaranteed Predictor
A lot of people discover xG and immediately think they’ve found a cheat code. They haven’t.
xG is a probability tool. It says “this is what should happen on average.” It doesn’t say what will happen today. A team with an xG of 2.0 might score zero goals if the keeper is on fire, or score four if a few half-chances somehow go in. That’s football. Variance is built into the game, and no model – however good – can remove it entirely.
So the right way to use xG is as one input among several, not as the only thing that matters.
Ignoring Context Behind the Numbers
Not all xG is created equal. A team that generates high xG by winning lots of penalties plays differently from a team that creates high xG through open-play combinations. Weather conditions, squad rotation, and tactical matchups all shape how well a team converts their expected goals in any given game.
And here’s a question worth asking before every bet: is the xG data I’m looking at recent, or is it pulled from a 38-game season average? Form matters. A team going through a defensive injury crisis in the last five weeks might have very different xG allowed numbers compared to their full-season stats. Always check the time frame of the data you’re using.
A Simple xG-Based Betting Framework
You don’t need a PhD to apply this. A basic workflow might look like:
- Check both teams’ average xG for and against over the last six to eight matches
- Note any big gaps between their xG and actual goals (over- or underperformers)
- Factor in head-to-head context – some teams historically suppress xG in big games
- Use the combined numbers to estimate a likely score range
- Compare that range to the odds available and look for value
This won’t win every bet. But it’ll probably stop you from placing bets based purely on “gut feeling” or recent hype around a team that just got lucky twice in a row.
Where to Find Reliable xG Data
Free and Accessible Sources
Several sites publish xG data for free. Understat covers the top five European leagues in solid detail. FBref (linked to StatsBomb data) goes even deeper, with shot-level breakdowns and xG per 90 minutes for individual players. For a quick pre-match check, even basic apps like Sofascore now show match xG after the final whistle.
The data is out there. Most bettors just don’t look for it.
Building a Simple Tracking Habit
Pick two or four leagues you regularly bet on. After each match day, record the xG numbers alongside the actual results. Within a month or two, you’ll start to see which teams consistently over- or underperform their expected goals, and that’s genuinely useful information.
It takes maybe ten minutes per week. And it’s probably the highest-value habit a casual football bettor can build without spending a penny.
xG won’t predict a last-minute deflected winner or a goalkeeper pulling off six saves in a row. But for bettors who want a smarter foundation for their predictions, expected goals explained in proper context give something most casual punters simply don’t have – a reality check on what the game actually showed, beyond whatever the scoreboard happened to say.
