Data has never been more important in soccer than it is today. It can tell you everything you want to know about an MLS team or a player, such as ground covered, passes attempted and completed, and duels won. It can also highlight upward and downward trends in performance.
Arguably the most often discussed data point is xG, or expected goals. This shows how many goals a team or player is expected to score across a period of time based on various factors. This sort of information is especially beneficial if you are betting on the game, or if you play in a fantasy league or anything like that. They even use it in post match analysis on TV shows these days. Most fans have heard of it.
It’s arguably most valuable to club scouts when analysing and recruiting players. A league like the MLS, which is in a growth phase and wants to attract star names as well as young up and coming talent, will look closely at metrics like xG. Major League Soccer is also home to many different playing styles, so xG provides a clearer picture of which teams are creating high-quality chances and which players are consistently getting into dangerous positions.
As a concept it is not difficult to understand, but as for how it is worked out, well, that’s a different matter. So I’m going to lay it all out for you in this article.
What is xG?
It stands for ‘expected goals’.
The xG is an ever-shifting metric that shows how likely a player is to score in multiple different scenarios.
xG is displayed as a figure between 0 and 1, reflecting the probability of a shot hitting the back of the net. So an xG of 0.5 suggests a 50% chance of a goal being scored given the situation at the time.
This information can be used to tell us all sorts of things about a team or a player.
Here are a few examples:
- If a team wins 1-0 but has a much lower xG for the game than their opponents, it suggests they were lucky to get the result.
- When a team consistently misses their xG, it might suggest they can create chances but their is an issue with finishing.
- A player consistently beating their xG could show they are an expert finisher, capable of converting chances most other players wouldn’t manage.
Another way of describing xG is as the quality of each chance. So a chance with an xG of 0.1 would be very difficult to score, while an xG of 1 would represent a tap in that even an old lady would struggle to miss.
So it’s not a literal prediction of how many goals should be scored, but a measure of the quality of each shooting opportunity, which therefore gives an indication of how well the shooter performs.
How is it Worked Out?
A few different companies compile sports data – Opta is arguably the best known. They all have their own methods of working out xG, but there are some aspects that are always included.
- Distance from goal – Closer shots from close to the goal have higher xG.
- Shot Angle – A shot directly in front of goal is easier to score than one from a tight angle, all other things being equal.
- Type of assist – A cut-back pass is better than a cross from deep, for example.
- Type of shot – A volley or header is harder to score than a shot on the ground, so this would lower the xG for that shot.
- Number of defenders nearby – More defenders result in a lower xG because they have more chance of intervening.
It’s also things like where the goalkeeper is, how much space the shooter has, how much pressure they are under from defenders, how the shooting opportunity came about (from a corner, from a counter attack, a long ball, etc).
They compile this data from hundreds of thousands of previous shots which allows them to come up with an xG score for every single shot taken. Since so many factors are included in every opportunity, the stats companies have built algorithms to create an xG score quickly. Data scientists input the information for each individual shot, and the algorithm does the rest.
The thing about xG is that it is based on historical data. So it can’t predict the future, it can only suggest what should happen based on what has happened in the past.
So if a striker has an xG of 0.8 in a game but scores twice, he has overperformed. If he doesn’t score at all he has underperformed. This is an over simplification on a game by game basis, but over a larger number of games the xG can highlight a player who does better than average in difficult situations, or who consistently misses chances he should have scored.