Xg stats predictions

The complete guide to xG football stat predictions

In today's climate with tipsters being in every nook and cranny and going off intuition is only going to take a few weeks before it will begin to bankrupt you.

Any profitable football betting pundit understands not only statistics but line movements and where a potential value will lie so he may bet at plus EV (expected value) margins.

Among the few stats a betting pundit relies on to analyze the underlying team's ability when creating chances the most valuable stat is XG or Expected Goals and taking luck and volatility out of the game is the aim for better tactical analyses of teams ability in front of goal.

Within this article we will lay down all there is to know in terms of calculating xG, converting it into profitable bets and implementing it into your regular weekly betting strategy.

Football stats: What is xG?

XG is a measure of how likely a team is to score a goal when presented with an opportunity. A goal can technically come from any shot but it is not the intention of the xG model to differentiate between these type of chances.

Any shot that may come to be on target or even just a shot taken in relation to another teams possession will be assigned a certain probability based on how many thousands of those specific types of chance are scored based on historically occurring records and patterns.

Henceforth, any shot given an xG value will always be between 0-1; a 0.10xG says out of hundreds of shots of that type the result is 10%. A penalty on the other hand would likely have a value between 0.76-0.78 depending on who is taking it.

ThexG value for a speculative 35 yard attempt taken by an attacking midfielder will rarely rise above 0.02.

4 Key components of the XG model:

In every modern day model there will always be individual metrics involved in calculating an xG value and there are in excess of 20 that may determine a shot to result in an xG value. Although all major providers are somewhat inconsistent in their analysis with Opta differing to StatsBomb and FBref the four fundamental ones for most sites include:

• The distance in yards/metres from the goal at which the shot was taken.

• The angle from which the shot was taken.

• What part of the body the shot was taken with.

• Which type of pass led to the shot, for example: a lay-off, a through ball or a cross.

• How many defenders were present at the point at which the shot was taken.

• Where the goalkeeper was standing when the ball was struck.

The origin of XG

The concept of the XG model is in itself not a modern concept as it's origins date back to American Football, Ice Hockey and Baseball.

Football however has been quite late on utilizing this statistical concept to the public eye thanks to Sam Green the prototype maker at Opta, and Michael Caley who later put the concept forward to a footballing audience by publishing the statistical calculation for football at Opta and FBref respectively.

Football is a very low scoring game, not only this, the statistical likelihood is that the weaker side who is the underdog throughout will be more than likely to end the game with 25 shots yet have a worse performance than their opponent and suffer the wrath of defeat via one of the opposition team taking one of their 2 shots and the goal bouncing off the defender or a keeper being out of position resulting in a cheap goal. XG can effectively erase these flukes.

The leading football stats prediction providers:

A competent football punter can readily identify the clear statistics to come out of a team's performance and convert those to an efficient betting strategy. Among the leading predictors include:

1. FBref/Opta. The two main providers of football statistics where one can get pretty much all that you need when considering xG/xGA in terms of calculations, the only setback is penalties being taken don't count toward xG/xGA calculations.

2. Understat. One of the best all-around betting statisticians I've seen so far, its stats platform is beautifully designed as well as providing helpful 'shot maps' and 'live odds' for when you are live betting games.

3. PredictZ/Infogol. With these two you won't need anything else to achieve a clear, simple betting outcome. Clear predictions of all available markets for the matches you wish to bet.

4. FootyStats/SoccerVista Extra. You can get detailed analysis of a team's xG trends over the last 10-20 games on both of these, and also get individual match breakdowns based on Home and Away stats.

2025/2026 Premier League xG and XGA

Taking an look at a few teams from last seasons Premier League campaign will help to solidify that there can indeed be an clear correlation between xG/xGA and the league standing of each club.

League Standing | Team | Expected Goals (xG) | Expected Goals Against (xGA) | Actual Points

---|---|---|---|---

1 | Arsenal (Champions) | 65.6 | 28.3 | 78

2 | Manchester City | 71.3 | 44.2 | 71

3 | Manchester United | 65.5 | 48.6 | 62

4 | Liverpool | 61.3 | 47.4 | 62

5 | Chelsea | 66.5 | 52.3 | 60

6 | Brighton | 57.9 | 49.1 | 58

7 | Bournemouth | 61.9 | 56.8 | 57

Analysis of 2025/2026 Season.

As one can tell from the league table Manchester City produced the most attacking xG that season with a whopping 71.3 but still did not emerge as league winners due to a lack of finishing from the striker(s) compared to Arsenal and the former side were less potent offensively than the league leaders in this particular campaign.

However, this is down to the fact that Arsenal was the 2nd highest for xGA with a figure of 28.3 showing how difficult it was to score against them.Chelsea on the other hand may well be one of the most frustrating teams to bet on for last seasons campaign and although they led the league in attacking xG value with 66.5 they were the 5th worse for xGA with 52.3 so it was always a chance of putting bets on markets such as BTTS and Over 2.5 Goals for all of Chelsea's league games.

How to bet on xGoals properly.

A lot of people think that xG is a shortcut to making money in betting however these claims will quickly be quashed once it begins to eat into your profit margins.

The key aspect for a football betting pundit is not just to analyze a teams xG for the upcoming match but to understand regression to the mean.

If one team had a xG of 5.2 during the last 5 games of a football season (for an example) yet they managed to score at 15 goals per game over that period, you will surely realize that they are performing above average and it's only a matter of time until that comes to a head and you can spot a valuable bet.

Frequently asked questions

1. What is the difference between xG and xGA?

XG represents the expected goals. XG is used to represent how well a teams attacking chances are. XGA represents expected goals against and refers to how good the chances a teams opposition has against them are. A good side will out score there opposition on xG with xGA.

2. Can an individual that is world class score more than their xG every time they play?

Yes. This is because elite players, those that finish clinicaly, such as Erling Haaland, Harry Kane and Kylian Mbapp, are just that much better than the average footballer and average values of xG are all based on an average player so therefore, a player who finishies clinically will most likely score more goals than they were expected to by their xG.

3. Why would xG numbers be different for different sites?

This is because they are all calculated differently and not every website models based on the same factors.

Therefore one site might give the shot a xG value of $0.45\text{ xG}$ because it accounts for the position of the goalkeeper while other sites will give the same chance a value of $0.35\text{ xG}$ if it does not account for the position of the goalkeeper.

Conclusion

In order to make gambling money reliable as rather than a guess, every single ticket you place must be calculated, and it will only be after goals become governed by xG and not the strikers that you will be able to take advantage of the bookmaker's inefficiencies in the odds.