Using xG to Analyze World Cup 2026 Matches: Beginner's Guide

What Is xG?

Expected Goals (xG) is a statistical metric that measures the quality of a goal-scoring chance by assigning a probability value between 0 and 1. A shot with an xG of 0.85 is one that a typical player would score roughly 85 times out of 100. By summing these values across all shots in a match, we arrive at a team’s total xG — a powerful indicator of how many goals they should have scored based on the chances they created.

How xG Is Calculated

Modern xG models are trained on thousands of historical shots and consider multiple variables:

• Shot location: Distance from goal and angle to the goalmouth are the strongest predictors of conversion probability.
• Body part: Headed shots are generally less likely to go in than footed shots from the same position, except in close-range scenarios.
• Assist type: Crosses, through balls, cutbacks, and passes from set pieces each carry different conversion baselines.
• Defensive pressure: Whether a defender is close by, whether the shot is taken under pressure, and the number of defenders between the shooter and the goal.
• Pre-assist movement: Some models account for the buildup play — a cutback after a dribble carries a higher xG than a hopeful long-range effort.

Each model provider (Opta, StatsBomb, Understat) uses slightly different weightings, which explains why you might see different xG values for the same match across platforms.

How to Read xG in Match Analysis

When analyzing a World Cup 2026 match, these xG benchmarks are useful:

A team that wins 2–0 but only generated 0.9 xG may have been lucky. A team that loses 1–0 but generated 2.4 xG was likely unfortunate and may perform better in future matches.

xG Leaders to Watch in 2026

Based on qualifying form and recent tournament data, these players historically outperform their xG (elite finishers) or generate high shot volumes:

• Kylian Mbappé: Consistently generates 0.5+ xG per 90 in international tournaments.
• Erling Haaland (if Norway qualifies): Elite shot volume with xG per 90 among the highest in world football.
• Lionel Messi: Creates high-quality chances from open play and set pieces alike.
• Harry Kane: Known for underrating his xG — his shot selection is excellent, often averaging 0.6+ xG per 90 at tournaments.

Team xG vs xGA for Defensive Analysis

Comparing a team’s xG (chances created) with their xGA (expected goals against) reveals their true performance profile:

• xG +2.0, xGA +0.4: Dominant attacking performance, excellent defensive structure.
• xG +0.8, xGA +2.2: Outplayed across the pitch — a win in this scenario is unsustainable.
• xG +1.5, xGA +1.4: An even contest despite what the scoreline may suggest.

At the 2022 World Cup, Morocco conceded just 0.8 xGA across their knockout matches, underscoring how their defensive organization was genuine rather than luck-driven.

Limitations of xG

xG is powerful but not perfect:

• No context on opposition: A high xG against a weak defense is less impressive than the same number against an elite backline.
• Ignores shot sequence: Two 0.1 xG shots from a corner routine that creates chaos are not the same as two isolated 0.1 chances.
• Model variance: Different providers assign different values to the same shot, making cross-platform comparison difficult.
• Goalkeeper quality: xG does not account for the specific goalkeeper facing the shot.

Application to Predictions

xG is one of the strongest single predictors of future match outcomes. A team with a consistently higher xG than their opponent over a 4–5 match sample is likely to progress, regardless of short-term results. For World Cup 2026, tracking xG differential across the group stage is an excellent way to identify teams that are underperforming (and due for regression) or overperforming (and likely to drop off in the knockouts).

Comparison with Actual Goals in Past World Cups

Data from the 2018 and 2022 World Cups shows that total xG closely tracks actual goals over a full tournament, but individual matches can vary wildly. In 2018, the average match xG was 2.6 against actual goals of 2.64 — remarkably close. In 2022, the average was 2.48 xG versus 2.53 actual goals. This convergence over a large sample size confirms xG’s reliability as a performance metric.