Problem with XG: Why the expected target is a defective metric

Expected Goals (XG) is everywhere in football analysis, presented as the main measure of performance. But really? Increased dependence on this single number hides significant problems: The problem with XG is that he is too simplifying a beautiful game, creating a deviant view of what really happened on the field. Although aiming to measure the quality of shots, the expected goal has some fundamental weaknesses that limit their use and can cause inaccurate conclusions about players and teams.

Here is the main problem XG that must be understood by every fan, analyst, and coach.

1. Statistical Problems: Small samples and high variance

The first big problem with XG lies in its statistical foundation. Football is a game with a low score where an ordinary shot only has a chance of around 10% to become a goal. This creates a very large statistical disorder. In order for player performance to be statistically different from random opportunities, they need to do a large amount of shots. Because most players do not reach this volume, the gap between the actual goals and their XG is often meaningless. This high variance makes it almost impossible to use XG to assess the player’s final completion skills reliably in a short time.

2. The problem of “blinding blindness”: XG ignores the player’s skill

Maybe the most striking problem with XG is “the blindness of the shooter”. The metric gives the same value to a shot, regardless of who took it. Opportunities that clearly have the same XG either fall into the hands of world -class strikers or central defenders. This is completely ignoring the reality of football, where the ability of final settlement varies dramatically from one player to another. The club pays millions of dollars for elite attackers precisely because they are better at converting opportunities than “average” players. By treating every shooter equally, XG fails to measure the skills that are often used to evaluate.

3. Context Problems: XG is in the vacuum space

Football matches are dynamic and complex, but XG analyzes moments separately. This creates significant context problems.

• This ignores the Buildup game: XG only measures its own shot, not a brilliant dribble or perfect bait that creates opportunities. Dangerous attacks foiled with the last tackle did not receive XG credit.

• This misses the game status: This metric does not take into account psychological factors such as pressure, momentum, or fatigue. The penalty in the last minute of the final match was treated the same as the penalty at the moment of the opening match of a friendly match.

• This ignores the goalkeeper: Most XG models do not take into account the quality or position of the goalkeeper. The high XG shot saved by the world -class goalkeeper is considered a failure, thus giving an unfair penalty to the performance of the shooter.

4. Team Quality Problems: Slanted results for the top teams

When you analyze XG for one full season, another problem arises. Top teams with elite attackers consistently scored more goals than their XG predictions. Conversely, teams that have difficulty scoring less scores often. This happens because the XG model is based on the “average” finisher, but the top team does not have an average finisher. This systematic bias means XG can be a bad tool to compare teams with different levels of quality, because XG consistently underestimates the best team efficiency and exaggerating the worst team potential.

5. Single Game Problem: Misuse XG for Match Analysis

One of the most common metric abuse is its application in a single game. The problem with XG is a tool to analyze long -term trends, not to assess the match 90 minutes. Preparation plays a big role in any game. States that a team “worth winning” because they have a higher XG is a mistake in the application of probabilistic metrics. High XG in defeat does not always mean a team of unlucky teams; This is just a very varied data point in sports.

Although the expected target can be a useful part of a broader analytical tool, these problems indicate why the expected target should not be a determinant of performance. The real story of a match is told in the context, skills, and drama of humanity-elements that can never be captured entirely by a single number.