xG in Football: What Expected Goals Mean and How to Use Them in Betting

A football score can be misleading. One team may take two shots and score twice, while the other creates ten dangerous chances and still fails to find the net. Looking only at the final 2–0 score can lead to completely wrong conclusions about how the match was played.

This is why football analytics uses xG — Expected Goals. The metric helps assess not how many goals a team scored, but how good its scoring chances actually were.

For betting analysis, xG is especially useful because it helps separate the underlying quality of a performance from finishing, randomness and individual match events.

What Is xG in Simple Terms?

xG estimates the probability that a particular shot will result in a goal.

Each shot is assigned a value between 0 and 1. The higher the number, the better the chance.

• xG 0.05 — roughly a 5% chance of scoring;
• xG 0.20 — around 20%;
• xG 0.50 — a very good chance with approximately a 50% probability of a goal;
• xG 0.80 — a chance that is converted in most similar situations.

The xG values of all a team's shots are added together.

For example, suppose a team creates four chances:

• 0.10 xG;
• 0.25 xG;
• 0.40 xG;
• 0.15 xG.

The total is:

This does not mean that the team "should have scored 0.9 goals." xG is a probability-based estimate of chance quality, not a prediction of an exact score.

How Is xG Calculated?

There is no single universal xG formula. Different data providers use their own models, so the same match may have slightly different xG figures depending on the source.

An xG model typically considers factors such as:

• distance from the shot to goal;
• shooting angle;
• which body part was used;
• the type of pass or assist before the shot;
• whether the chance came from open play, a corner or a free kick;
• the position of defenders;
• the goalkeeper's position;
• the type of attack;
• whether the shot was taken with the foot or the head.

The model compares the chance with a large number of similar situations from previous matches and estimates how often those chances resulted in goals.

A shot from almost on the goal line will therefore receive a high xG value, while a speculative effort from 30 metres will usually receive a very low one.

Why Can the Score and xG Be So Different?

Imagine a match that ends 3–1.

At first glance, it may appear that the winning team comfortably dominated. But the xG numbers show:

That tells a very different story.

Team A converted almost everything it created, while Team B generated better chances overall but finished them less efficiently.

This does not mean that the result was somehow "wrong." Goals still decide the match. But when analysing future fixtures, it is useful to understand whether the final score accurately reflected the quality of the chances created.

What Are xG Overperformance and Underperformance?

If a team consistently scores noticeably more goals than its xG would suggest, it is said to be overperforming its xG.

For example:

The team has scored almost six goals more than its xG would suggest.

The opposite situation might look like this:

Here, the team is creating chances but converting them less efficiently than expected.

However, it would be wrong to assume that the team will automatically start scoring more simply because it has underperformed xG. The gap between goals and xG may be caused not only by randomness, but also by player quality, the limitations of the model and the team's style of play.

What Is xGA?

If xG measures the quality of chances a team creates, xGA (Expected Goals Against) measures the quality of chances it allows its opponents to create.

For example:

• team xG — 1.85;
• xGA — 0.92.

This suggests that the team is creating considerably better chances than it is allowing at the other end.

Over a meaningful sample, the difference between xG and xGA can help assess the balance between a team's attacking and defensive performance.

How to Use xG When Analysing Totals

One of the most obvious uses of xG is in over/under markets.

Suppose two teams have repeatedly played Under 2.5 Goals in their recent matches. Based only on results, both may appear to be low-scoring teams.

But then you find that:

• the first team averages 1.8 xG;
• the second averages 1.6 xG;
• both regularly score fewer goals than might be expected from the quality of their chances.

In that situation, the recent run of low-scoring matches becomes a less convincing argument on its own.

The reverse is also true. A run of matches with four or five goals may come with fairly modest xG figures, suggesting that unusually efficient finishing contributed to the high scores.

If you want to understand these markets in more detail, see our guide to totals and handicaps in sports betting.

xG for Match Results and Handicap Betting

xG is useful for more than just totals.

Suppose a team has lost three matches in a row. Its recent form looks terrible:

But the xG numbers from those matches were:

The results are poor, but the quality of the chances created suggests that the team's performances were significantly better than the losing streak alone would indicate.

This kind of information can be useful when analysing odds for match winners, double chance or handicap markets.

You can read more about positive and negative handicaps in our separate guide: "What Is a Handicap in Betting?".

Why You Should Not Judge a Team by One Match's xG

One of the biggest mistakes is treating a single-match xG figure as a final verdict.

Football is too variable for that.

A team may receive a red card after 15 minutes. It may score early and deliberately surrender possession. It may be facing a much stronger opponent. Or it may create one chance worth 0.9 xG and very little else.

For that reason, xG is much more useful over a longer period.

The same principle applies to almost any betting statistic: a small sample can be highly misleading.

We explain this in more detail in "Sample Size in Betting: Why a Few Matches Prove Almost Nothing".

xG Does Not Eliminate Variance

Even a chance with 0.70 xG does not have to result in a goal. A 70% probability also means that around 30% of similar chances are not converted.

The same works in reverse: a shot with 0.03 xG will occasionally fly into the top corner.

This is why a team can create the better chances and still lose. Another team can produce almost nothing and win because of one exceptional long-range strike.

That is a normal part of football.

The relationship between probability, randomness and results is covered in more detail in our article about variance in sports betting.

What xG Does Not Tell You

xG is useful, but it should not be treated as a universal answer to every football question.

It does not capture the entire match.

For example, xG may not fully reflect:

• dangerous attacks that never result in a shot;
• a final pass that the striker narrowly fails to reach;
• the tactical effect of an early goal;
• individual finishing ability;
• differences between xG models used by different data providers;
• changes in the game after a red card;
• motivation and tournament context.

The first point is particularly important.

If a striker receives a cross in front of an open goal but misses the ball by a few centimetres, no shot is recorded. As a result, a standard xG model may not record this very dangerous situation at all.

Common Mistakes When Using xG

Mistake 1. "The Team Had Higher xG, So It Deserved to Win"

xG measures shot quality. It does not determine a "fair" winner.

Mistake 2. "The Team Is Five Goals Below xG, So It Will Score Five Extra Goals Soon"

Probability does not work that way. Underperformance in previous matches does not mean a team must make up the difference in the next ones.

Mistake 3. Looking Only at the Last Two or Three Matches

Over such a short period, the numbers can be heavily influenced by the specific opponents and match situations.

Mistake 4. Treating xG From Different Providers as Identical

xG models differ. For consistent analysis, it is usually better to use the same data source over time.

Mistake 5. Using xG Without Looking at the Odds

Even a strong statistical idea does not automatically make a bet valuable. You still need to consider the probability already reflected in the bookmaker's price.

How to Analyse a Match Using xG

A practical process might look like this:

1. Check both teams' xG and xGA over a meaningful period.
2. Separate home and away matches if the sample is large enough.
3. Look beyond the average and consider the opposition. Producing 2.0 xG against weak teams is not the same as producing 2.0 xG against title contenders.
4. Compare actual goals with xG. A large difference may indicate unusually high or low finishing efficiency, but it does not automatically mean results will revert to the average in the next few matches.
5. Consider team news. Missing a key striker or several defenders can significantly change expectations.
6. Consider the match context. League position, the need to win and whether a draw suits both teams may all matter.
7. Only then compare your assessment with the bookmaker's odds.

Conclusion

xG is one of the most useful metrics in modern football analysis. It allows you to look beyond the final score and assess the quality of the chances a team actually creates and concedes.

However, xG should not be used as a ready-made prediction.

It works best alongside xGA, team news, opponent strength, home and away performance, playing style and bookmaker odds.

Put simply, the score shows the result, while xG helps explain how closely that result matched the quality of the chances created. That is why xG can often provide more useful information for analysing future matches than the recent results table alone.