Expected Goals, What xG Can and Cannot Tell You
xG estimates the likelihood of a shot becoming a goal; it does not say that a goal was guaranteed.
Expected goals, or xG, estimates a shot's scoring probability from the characteristics used by a model. FIFA defines the metric as the average probability of scoring from an individual attempt. It therefore answers a different question from shot totals or actual goals.
A hypothetical calculation
A shot valued at 0.20 has an estimated probability of about twenty per cent within that model. It does not promise exactly one goal in the next five similar attempts. Three shots valued at 0.10, 0.20 and 0.40 produce a total of 0.70 xG.
That total does not mean a seventy per cent probability of scoring at least once. It is a sum of expected values, not that probability calculation. Scoring a low-value chance and missing a higher-value one is also entirely compatible with the model.
What the total reveals
xG helps distinguish numerous difficult shots from fewer, more favourable opportunities. It does not capture every dangerous attack that ends before an attempt. An intercepted final pass may reveal a significant attacking problem without adding anything to the shot-based total.
Keep comparisons consistent
Models can use different inputs, so compare figures from the same provider rather than treating all published numbers as interchangeable. Check whether penalties are included as well.
Read xG alongside the timing of chances, the score at the time and the match itself. One game is a small sample for judging a striker or goalkeeper. A team total also cannot explain who created the space or made the best decision before the shot.
Share this article
