In the football world, data and statistics are becoming increasingly important. Fans rarely focus solely on the score. Instead, they look into details like possession percentages, shots on goal, and more recently, advanced metrics such as xG. But how is this metric calculated, and how can it be used for further analysis?
What is xG? What are Expected Goals?
Expected Goals (xG) measure the quality of a scoring chance by analyzing similar shots from the past. Each shot is assigned an xG value between zero and one. Zero represents a virtually impossible chance to score, and one indicates a chance that would result in a goal 100% of the time. In reality, most chances fall between 0.01 and 0.99, since nothing in football is ever completely certain.
Simply put, if a player takes a shot with an xG value of 0.1, it means that, on average, there’s a 10% probability of scoring from that position.
How is xG calculated?
Several factors contribute to a shot’s xG value. Key considerations include the distance to goal, the angle of the shot, the number of defenders between the shooter and the goal, the type of pass that set up the chance, and other variables that can differ depending on the xG model. Additionally, the type of shot is crucial, for example, it’s generally easier to score with a shot taken from the ground using the foot than with a volley or header. Thousands of similar shots are analyzed to determine the likelihood of a goal in each situation. More recently, models have even started to incorporate the goalkeeper’s position and line of sight, since a keeper with an obstructed view has less time to react.
What are the benefits of using xG?
Now that we understand what xG is and how it’s calculated, what can we do with this information? The real advantage of xG lies in its ability to provide a deeper insight into a game’s outcome beyond traditional statistics like possession, shots and passes. By comparing a team’s or player’s actual goals to their xG, you can gauge how effective they were in converting their chances. For instance, if a team scores more goals than their xG suggests, it might indicate exceptional finishing, or simply a touch of luck.
Moreover, this metric helps shed light on overall team performance. If a team goes through a rough patch, looking at the xG values might reveal that they’re simply underperforming both offensively and defensively over the last few games. This approach is also key to our strength ratings, where xG plays a major role in quantifying team quality. That said, some people take xG too literally. Let’s explore some common misconceptions regarding Expected Goals.
Common Misconceptions
Often, after a loss, fans will say, “We should have won; we had way more expected goals.” While there is some truth behind that sentiment, xG shouldn’t be taken out of context to explain a single game. For example, if Team 1 takes 15 shots with an average xG of 0.1 each but scores no goals, while Team 2 shoots less frequently yet scores 2 goals from opportunities averaging 0.6 xG, Team 1 may have a higher overall xG but wasn’t necessarily the better team.
That’s why, when analyzing Expected Goals, it’s important to consider a broader perspective, such as an entire season or even a player’s career. Some players may experience short-term dips where they underperform their xG stats, but reviewing multiple seasons can reveal just how crucial they are in front of goal.
Take Messi, for example, arguably the greatest player of all time. Even though he underperformed his Expected Goals during his time at PSG, the bigger picture tells a different story. Since the start of the 2017/18 season, he has overperformed his Expected Goals by a staggering 35 goals. Ridiculous. This illustrates why we shouldn’t overemphasize xG stats over short periods; instead, it’s best to give the data time to show the true trends.