Expected Goals (xG) Calculator & Converter

In modern football betting, Expected Goals (xG) is one of the most widely used metrics for evaluating team performance. It estimates how many goals a team should have scored based on the quality of chances created, filtering out some of the noise from luck and variance.

However, a raw xG number is not directly actionable. Knowing that “Liverpool had 2.45 xG” does not tell you whether to back them or not. Our xG Calculator bridges that gap — converting xG inputs into 1X2 Probabilities, Over/Under, BTTS, Fair Odds, and a Correct Score Heatmap using the Poisson distribution. You can also compare the calculator’s Fair Odds with your bookmaker’s prices to check for potential value.

How to Use the xG Calculator

This tool converts Expected Goals into baseline probabilities for the most common football betting markets. Follow these steps:

1. Find the Projected xG: Get the xG estimate for the upcoming match. Good free sources include FBref (StatsBomb data), Understat, and Infogol.
   • Important: Input projected match xG — an estimate of what each team is expected to produce in this specific fixture — not a raw season average. A season average ignores opponent strength, home/away split, and recent form. For the best results, use a weighted combination of each team’s xG For and xG Against, adjusted for the opponent.
2. Enter the Numbers: Input the Home xG and Away xG. Default values are provided as a starting point.
3. (Optional) Enable Dixon-Coles Correction: Check the toggle to apply a low-score correlation adjustment. This increases the probability of draws (0-0, 1-0, 0-1, 1-1) to better match observed football data. Recommended for matches where both xG values are below 1.5.
4. Read the Results:
   • 1X2 Cards: Probability and Fair Odds for Home Win, Draw, Away Win.
   • Bookmaker Comparison: Enter your bookmaker’s odds under each card. The calculator flags “Value” (green) when the bookmaker’s odds exceed the Fair Odds, and “No Value” (red) when they don’t.
   • Over/Under: Probabilities and Fair Odds for O/U 0.5 through 4.5 total goals.
   • BTTS & Clean Sheet: Both Teams To Score (Yes/No) and individual Clean Sheet probabilities.
   • Correct Score Heatmap: A 6×6 grid showing the probability of each scoreline from 0-0 to 5-5. Darker green = higher probability.

Related Tools: If the xG data suggests a tight match with a high draw chance, consider our Draw No Bet Calculator. If the model predicts many goals, verify with our Over/Under Goals Calculator. For a heuristic pre-check based on form rather than xG, see the Match Analysis Calculator.

The Model: Independent Poisson Distribution

The calculator uses the Independent Poisson Model, one of the most established baseline approaches in football analytics.

How it works: Given a team’s expected goals (λ), the Poisson formula calculates the probability of scoring exactly k goals:

P(k) = (λ^k × e^−λ) / k!

The calculator computes this for both teams independently (Home goals 0–9 and Away goals 0–9), generating a 10×10 matrix of scoreline probabilities. These are then summed to produce 1X2 outcomes, Over/Under totals, and BTTS.

Dixon-Coles Correction (Optional)

The standard Poisson model treats Home and Away goals as independent events. In reality, there is a small negative correlation — low-scoring draws (0-0, 1-1) occur slightly more often than independent Poisson predicts, partly because teams sometimes reduce risk when the score is level. The Dixon-Coles adjustment applies a correction factor (ρ ≈ −0.13) to the four lowest scorelines (0-0, 1-0, 0-1, 1-1), then renormalizes the full matrix. This tends to increase Draw probability by 1–3% and decrease extreme scoreline probabilities slightly.

Limitations of This Model

No model is perfect. Here is what this calculator does not account for:

• Goal correlation beyond Dixon-Coles: Even with the correction, the model does not dynamically adjust for game state (e.g., a team trailing will press harder, changing both teams’ goal rates).
• Individual finishing quality: xG is calibrated to the average player. A team with an elite finisher (historically overperforming xG) or an elite goalkeeper (underperforming xGA) may deviate systematically. This is sometimes called “xG over/underperformance.”
• Team news and context: Injuries, suspensions, rotation, weather, and tactical changes are not reflected in a two-number xG input.
• Market information: Bookmaker odds aggregate vast amounts of information (insider knowledge, sharp money, public opinion). The model’s Fair Odds are a mathematical baseline, not a replacement for market prices.

Use the calculator’s output as a starting point for analysis — compare it with market odds to identify discrepancies worth investigating further.

Descriptive xG vs. Projected xG: What Should I Enter?

This is one of the most common mistakes with xG calculators.

Descriptive xG is the xG from a match that has already been played. It tells you how the game went — useful for post-match analysis, but not for predicting the next match.

Projected xG is an estimate of what a team will produce in a future fixture, accounting for opponent strength, venue, and recent form. This is what the calculator needs.

If you simply enter a team’s season-average xG, you ignore the specific opponent. A team averaging 1.80 xG per game will not produce 1.80 xG against every opponent — they might expect 2.30 against a weak defense and 1.10 against a top-four side. The more accurately you estimate the projected xG, the more useful the calculator becomes.

Real-World Examples

Example 1: Identifying an Overpriced Favourite

Manchester United are playing Crystal Palace. The bookmaker prices United at 1.50 (implied 66.7%).

• The Data: Based on recent form and opponent-adjusted xG, you estimate United at 1.20 xG and Palace at 1.10 xG.
• The Calculation: The Poisson model gives P(Home Win) ≈ 38.5%, Fair Odds ≈ 2.60. P(Draw) ≈ 27.5%. P(Away Win) ≈ 34.0%.
• The Comparison: The bookmaker’s 1.50 implies 66.7% — nearly double the model’s 38.5%. The model strongly suggests this is poor value. You enter 1.50 in the Bookmaker Odds field and see a red “No Value” badge.
• Note: This does not mean United will lose. It means that at these xG inputs, the bookmaker’s price overstates United’s chances according to the model. If your xG estimate is wrong, the conclusion changes.

Example 2: Using the Heatmap for Correct Score

Arsenal vs Brighton. You estimate Arsenal xG: 2.15, Brighton xG: 0.85.

• The Heatmap: The most probable scorelines are 2-0 (≈17.4%), 1-0 (≈16.2%), 2-1 (≈13.8%), 1-1 (≈11.7%). The heatmap shows 2-0 and 1-0 as the darkest cells.
• Derived Markets: Over 2.5 Goals ≈ 47%, BTTS Yes ≈ 47%. The model suggests this is a match that could go either way on total goals — not a clear Over or Under play.

Frequently Asked Questions (FAQ)