In short
The answer first
Expected Goals, usually written xG, estimates the probability that a shot will become a goal based on the characteristics of that chance and a model trained on historical shots. A higher xG value represents a better scoring chance, but it is a probability estimate—not a statement that a particular shot should definitely be scored or missed.
Key takeaways
xG evaluates the quality of shots rather than simply counting them.
Common model inputs include shot location, angle, body part and the action that created the shot, with richer models using more context.
Team xG is usually the sum of the xG values assigned to its shots.
Different data providers can give the same shot different xG values because their models and data inputs differ.
xG is most useful as context alongside video, shot volume, game state and other evidence—not as a replacement for the actual score.
What the number is trying to estimate
An xG model asks how often shots with similar characteristics have historically become goals. A close-range chance with a clear angle will usually receive a much higher value than a speculative shot from long distance because comparable chances have been converted more often.
The number is normally expressed between zero and one. A value around 0.20 can be read as roughly a 20 percent modelled scoring probability under that provider's methodology—not a guarantee that exactly one in five identical-looking shots will be scored in every small sample.
How a model builds shot quality
Opta describes variables such as assist type, shot angle, distance from goal, whether the attempt was a header and whether it meets its big-chance definition. Other models may include pressure, goalkeeper location, preceding actions and additional spatial context.
Because the models have different data and design choices, there is no single universal xG value that football's Laws define as the correct number for a shot.
Reading team and player xG
Adding the shot values gives a summary of the chances a team or player generated. A team can lose 1–0 while producing more xG, which may suggest it created the stronger set of shooting opportunities without converting them on that day.
That does not make the real result wrong. Goals decide matches; xG helps analyse the process behind those goals and misses. Over longer samples, analysts can compare goals with xG to study finishing, chance creation and defensive shot quality.
What xG does not tell you by itself
One total cannot fully describe a match. Two teams can each post 1.5 xG with very different chance profiles: one may create a single huge opportunity plus small shots, while the other creates several medium-quality chances.
Game state also matters. A team protecting a lead may willingly concede harmless possession and low-quality shots. The best use of xG is therefore to start a better question about chance quality, not to end every tactical discussion with one number.
Source check
Sources and further reading
KickVerse uses primary governing-body material where a rule or competition format has an official source. Analysis terms can vary between data providers, so provider-specific definitions are identified rather than presented as universal laws.
Continue learning
Related KickVerse guides that add useful context to this topic.