WHY XG IS FLAWED
WHY XG IS FLAWED
Goals Above Expected, most commonly known as xG, has been all the rage in soccer analytics in recent years. It has given fans and pundits a measurable way to evaluate a team's chances of scoring and to identify players who are underperforming or overperforming their expected goals. However, as with any metric, there are limitations and potential flaws to consider when using xG.
The Flawed Assumptions Behind xG
At its core xG is a model that relies heavily on historical data and statistical analysis to create an 'expected' outcome from a given shot. The problem is, however, that soccer is a chaotic game, and it is impossible to account for every variable that can affect a shot's outcome (e.g. individual player quality, team strategy, zonal marking, low-block defences). No matter how much data is available or how sophisticated the model is, there will always be a level of uncertainty surrounding where a shot is taken from.
Inevitable Omissions in the XG Calculation
Another limitation of xG is that it only considers the final shot taken, ignoring the quality of the chances created leading up to it. A player might create multiple chances per game, but only the shots taken will be counted towards their xG tally. This can be misleading as it fails to acknowledge the player's overall contribution to the team's attacking play. Similarly, xG does not take into account the quality of the opposition's defense, which can significantly impact the likelihood of a goal being scored.
Oversimplification of Complex Interactions
xG also has difficulty accounting for the dynamic interactions between players on the pitch. For example, a player might have a low xG because they often take shots from difficult angles or because they are often forced to play through a crowded defense. However, this does not necessarily mean that they are a poor player; it could simply indicate that they are playing in a system that does not suit their strengths.
Hypothetical Scenarios vs. Real-life Context
A major flaw of xG is that it is based on hypothetical scenarios rather than real-life context. The model assumes that every shot has an equal chance of being scored, regardless of the circumstances surrounding the shot. This is simply not true in practice. For example, a player might have a low xG because they often take shots from long range or because they are often forced to play against a team with a strong defense. In these cases, the player's xG may not be an accurate reflection of their true ability.
Not Accounting for Finishing Ability
Finally, xG does not take into account the finishing ability of the player taking the shot. Some players are simply better finishers than others, and this can have a significant impact on the likelihood of a goal being scored. A player with a high xG may not be as dangerous as a player with a lower xG if the latter is a more clinical finisher.
Conclusion
xG is a useful tool that can provide insights into a team's or player's attacking performance. However, it is important to remember that it is just a model, and it has limitations. It is important to use xG in conjunction with other metrics and scouting reports to get a more complete picture of a player's or team's performance.
FAQs
Q: Why is xG flawed?
A: xG is flawed because it relies on historical data and statistical analysis, which may not always be accurate. It also does not account for the quality of the chances created leading up to a shot, the quality of the opposition's defense, or the finishing ability of the player taking the shot.
Q: What are the limitations of xG?
A: xG only considers the final shot taken, ignoring the quality of the chances created leading up to it. It also does not take into account the quality of the opposition's defense or the finishing ability of the player taking the shot.
Q: Can xG be used to evaluate a player's or team's performance?
A: xG can be used to provide insights into a team's or player's attacking performance, but it should not be used as the sole metric for evaluation. It is important to use xG in conjunction with other metrics and scouting reports to get a more complete picture of a player's or team's performance.
Q: What are some alternatives to xG?
A: Some alternatives to xG include shot-based metrics such as shots on target and shots on goal, as well as non-shot-based metrics such as expected assists and pass completion percentage.
Q: How can xG be improved?
A: xG can be improved by incorporating more data into the model, such as data on the quality of the chances created leading up to a shot, the quality of the opposition's defense, and the finishing ability of the player taking the shot.
Leave a Reply