Expected Assists (xA) is an advanced football statistic that measures the quality of chances created by a player through their passes, crosses, and through balls. While expected goals (xG) evaluates the quality of shots taken, xA evaluates the quality of the final pass that sets up a shot. Each key pass — defined as a pass that directly leads to a shot — is assigned an xA value equal to the xG of the resulting shot. If a player delivers a cross that leads to a header with an xG of 0.15, that cross receives an xA value of 0.15. A player’s cumulative xA over a match or season represents the total goal probability they have created for teammates through their passing.
How xA Differs from Traditional Assists
Traditional assists are a binary statistic — a player either gets an assist (when their pass directly precedes a goal) or they do not. This binary measurement fails to capture the full picture of creative output because it depends not only on the quality of the chance created but also on whether the teammate finishes it. A player who delivers ten perfect through balls that result in one-on-one situations with the goalkeeper but whose teammates squander every one receives zero assists. Under the xA framework, the same player would accumulate a substantial xA total reflecting the genuine quality of chances they created, regardless of their teammates’ finishing.
This distinction is crucial for player evaluation. A midfielder playing for a team with poor finishers may have a low assist count but a high xA, indicating that their creative output is excellent even though the final product is let down by teammates. Conversely, a player with a high assist count but low xA may be benefiting from teammates who convert low-quality chances through individual brilliance. By separating creation quality from finishing quality, xA provides a more accurate assessment of a player’s true creative ability.
The gap between a player’s actual assists and their xA (sometimes called the assist luck component) tends to regress over time, similar to the regression between goals and xG. Players who significantly overperform their xA in one season — recording many more assists than their xA suggests — tend to see their assist numbers decline in subsequent seasons. Players who underperform their xA tend to see improvements. This regression pattern makes xA a more reliable predictor of future assist output than historical assist counts alone.
The top xA accumulators in European football are typically creative midfielders and attacking full-backs who deliver high volumes of key passes in dangerous areas. Players like Kevin De Bruyne, Trent Alexander-Arnold, and Thomas Muller consistently rank among the highest xA generators because they combine exceptional passing ability with tactical positions that involve frequent delivery into the penalty area. Their high xA reflects genuine creative talent that produces value for their team regardless of the finishing quality of the specific teammates they play alongside.
xA in Team Analysis and Match Prediction
At the team level, xA data helps identify which teams create the highest-quality chances and through which mechanisms. A team with high xA from crosses suggests a wide-play dependent attack, while high xA from through balls indicates a team that penetrates through central channels. This stylistic information is valuable for match prediction because it helps assess how a team’s creative approach will interact with the specific defensive strengths and weaknesses of their opponent.
When a team faces an opponent that is strong against crosses but vulnerable to through balls, the team’s xA from through balls becomes more relevant than their overall xA. Similarly, when a team loses its primary creative player to injury, the reduction in expected chance quality should be factored into match predictions even if the team’s recent results have been unaffected. xA data provides the granularity needed for these detailed tactical assessments.
For correct score prediction, xA contributes to the estimation of each team’s expected goals by measuring the supply side of chance creation. A team’s xG is ultimately a function of both the quantity of shots they take and the quality of those shots, and xA helps explain why certain teams generate higher-quality chances than others. At Correct Score Predict, creative output data including xA informs our assessment of each team’s attacking quality, helping produce more accurate scoreline forecasts.
xA and Betting Applications
xA data has practical applications across several betting markets. In the Anytime Goalscorer market, identifying which players receive the most high-xA service helps predict who is most likely to score. A striker who receives passes with high cumulative xA from creative teammates is effectively being given more and better chances to score, which should be reflected in their goalscorer probability even if their recent finishing has been below par.
For assist-specific betting markets — offered by some bookmakers as part of player performance bets — xA provides a much better basis for prediction than historical assist counts. A player’s xA per 90 minutes is a more stable and predictive metric than their assists per 90, making it the preferred input for any model aimed at predicting future assist output. At Correct Score Predict, our analysis considers the full range of advanced metrics including xA to provide comprehensive match forecasts that capture both the scoring and creative dimensions of team performance.
