xGA and xGD explained: measuring a team's real quality
Expected goals against and expected goal difference in plain terms, how they measure defence and overall control, why they predict future results better than the league table, and how to use them.
The idea in one sentence
Expected goals against, or xGA, is the flip side of expected goals: it adds up the quality of the chances a team allows its opponents, giving a measure of how well it defends. Expected goal difference, or xGD, is simply a team's expected goals minus its expected goals against. Together they describe how much control a side has over a match — how many good chances it creates versus how many it concedes.
How they are built
Both numbers come from the same chance-by-chance valuation that produces expected goals. Every shot a team faces is rated by how likely it was to be scored, and the sum is the team's xGA. Subtract that from the xG the team generated at the other end and you have its xGD for the match or the season. A positive xGD means a team is creating better chances than it allows; a negative one means the opposite, whatever the scoreboard happens to say.
Why they beat the table
Actual results are noisy. A team can win on a deflection or lose to a wonder-strike, and goalkeeping and finishing streaks swing scorelines in ways that do not last. xGD smooths this out by measuring the underlying quality of play. Over a season it correlates more closely with future results than actual goal difference does, because it reflects what a team repeatedly does, rather than the handful of moments that happened to fall for or against it.
How to read it sensibly
The most useful signal is the gap between a team's real results and its expected numbers. A side sitting high in the table but with a modest xGD may be overperforming and due to fall back, while one stuck in mid-table with a strong xGD is often playing better than its points suggest and likely to climb. As always, more matches make the picture more reliable, and context such as game state and red cards should temper any single reading.
Where you will see it
xGA and xGD appear in season previews, form guides and the kind of analysis that tries to see past a lucky or unlucky run of results. They are central to how models estimate a team's true strength, and they feed directly into match predictions: a side with a strong underlying xGD is judged more capable than its recent scoreline suggests, which is exactly the sort of edge a data-driven forecast is built to find.