The Problem with xG: Why Expected Goals is a Flawed Metric

Expected Goals (xG) is everywhere in football analysis, presented as the ultimate measure of performance. But is it? The growing reliance on this single number masks a significant issue: the problem with xG is that it oversimplifies the beautiful game, creating a distorted view of what truly happens on the pitch. While it aims to quantify shot quality, Expected Goals has several fundamental flaws that limit its use and can lead to inaccurate conclusions about players and teams.

Here are the key problems with xG that every fan, analyst, and coach should understand.

1. The Statistical Problem: Small Samples and High Variance

The first major problem with xG lies in its statistical foundation. Football is a low-scoring game where a typical shot has only about a 10% chance of becoming a goal. This creates massive statistical noise. For a player’s performance to be statistically different from random chance, they would need to take an enormous number of shots. Since most players don’t reach this volume, the gap between their actual goals and their xG is often meaningless. This high variance makes it nearly impossible to use xG to reliably judge a player’s finishing skill over a short period.

2. The “Shooter Blindness” Problem: xG Ignores Player Skill

Perhaps the most glaring problem with xG is its “shooter blindness.” The metric assigns the same value to a shot regardless of who is taking it. A clear chance has the same xG whether it falls to a world-class striker or a center-back. This completely ignores the reality of football, where finishing ability varies dramatically from player to player. Clubs pay millions for elite forwards precisely because they are better at converting chances than an “average” player. By treating every shooter the same, xG fails to measure the very skill it is often used to evaluate.

3. The Context Problem: xG Exists in a Vacuum

Football matches are dynamic and complex, but xG analyzes moments in isolation. This creates a significant context problem.

• It Ignores Buildup Play: xG only measures the shot itself, not the brilliant dribble or perfect pass that created the chance. Dangerous attacks that are thwarted by a last-ditch tackle receive zero xG credit.

• It Misses Game State: The metric does not account for psychological factors like pressure, momentum, or fatigue. A penalty in the last minute of a final is treated the same as one in the opening moments of a friendly.

• It Overlooks Goalkeeping: Most xG models do not factor in the quality or positioning of the goalkeeper. A high-xG shot saved by world-class goalkeeping is simply registered as a miss, unfairly penalizing the shooter’s performance.

4. The Team Quality Problem: Skewed Results for Top Teams

When you analyze xG over a full season, another problem emerges. Top teams with elite attackers consistently score more goals than their xG predicts. Conversely, struggling teams often score fewer. This happens because the xG model is based on an “average” finisher, but top teams don’t have average finishers. This systematic bias means xG can be a poor tool for comparing teams of different quality levels, as it consistently underestimates the efficiency of the best and overestimates the potential of the worst.

5. The Single-Game Problem: Misusing xG for Match Analysis

One of the most common misuses of the metric is applying it to single games. The problem with xG is that it is a tool for analyzing long-term trends, not for judging a 90-minute match. Randomness plays a huge role in any one game. Declaring that a team “deserved to win” because they had a higher xG is a fundamental misapplication of a probabilistic metric. A high xG in a loss doesn’t necessarily mean a team was unlucky; it’s simply a data point in a highly variable sport.

While Expected Goals can be a useful part of a broader analytical toolkit, these problems show why it should never be the final word on performance. The real story of a match is told in its context, skill, and human drama—elements that a single number can never fully capture.