Expected Goals (xG) and Its Role in Predicting Correct Scores

What is Expected Goals (xG)?

Expected Goals (xG) is a statistical metric that evaluates the quality of goal-scoring chances. It assigns a value between 0 and 1 to every shot, reflecting the probability of that shot resulting in a goal.

Examples:

A penalty = ~0.76 xG
A shot from outside the box = ~0.02 xG
A close-range header = ~0.30 xG

These values are based on historical data, considering factors like:

Shot location
Type of assist
Body part used (foot, head)
Defensive pressure
Game context (open play, set piece, etc.)

Why xG Matters for Correct Score Predictions

Traditional metrics like goals scored or conceded can be misleading. xG helps you see beyond the final score and evaluate how well a team actually performed.

Example:
Team A 1–0 Team B
xG: Team A 0.45 – 1.70 Team B
This tells you Team B was unlucky not to score — next time, you might lean toward a different prediction based on underlying performance.

How to Use xG in Correct Score Models

Here’s how xG can enhance your correct score predictions:

1. Calculate Attack and Defense Strengths Using xG

Instead of using goals per game, calculate:

Attack Strength = Team xG scored / League average xG scored
Defense Strength = Team xG conceded / League average xG conceded

2. Adjust Team Ratings Based on xG Trends

Use rolling averages (e.g., last 5 matches) to spot form changes:

A team underperforming xG may be “due” a bigger scoreline.
A team overperforming might regress to the mean.

Feed xG into Poisson Models

The Poisson distribution requires expected goals for both teams. xG provides a better foundation than raw goals.

Formula:
Adjusted attack × opponent’s defense × league avg goals = Expected Goals for match

Use these to calculate probabilities for each scoreline (0-0, 1-0, 2-1, etc.).

Real-World Example

Let’s say:

Manchester City: xG for = 2.4/game, xG against = 0.8/game
Burnley: xG for = 0.9/game, xG against = 2.1/game
League average = 1.5 xG/game

Man City Expected Goals = (2.4 / 1.5) × (2.1 / 1.5) × 1.5 ≈ 2.24
Burnley Expected Goals = (0.9 / 1.5) × (0.8 / 1.5) × 1.5 ≈ 0.48

Use a Poisson distribution to calculate:
Most likely scorelines: 2-0, 2-1, 3-0

Limitations of xG

While xG is powerful, it’s not perfect:

Doesn’t account for post-shot quality (e.g. goalkeeper saves)
Ignores psychological or situational factors
Doesn’t measure finishing skill or clinical strikers

Combine xG with qualitative insights like form, motivation, injuries, and tactics for optimal accuracy.

Explore xG Stats by League

Dive deeper into Expected Goals data across the top European leagues:

Conclusion

Expected Goals (xG) has revolutionized the way analysts and bettors assess football performance. When applied correctly, xG:

Improves your understanding of team strengths
Helps build more accurate models for correct score predictions
Reduces reliance on misleading raw scorelines

If you're serious about predicting football scores — especially correct ones — xG isn't optional. It’s your analytical foundation.