Poisson Distribution for Football Betting: Build Your Own Model
Learn how to use the Poisson distribution to predict football match outcomes. Step-by-step guide to building a statistical model for scoreline and goals predictions.
Editorial Team
Published 14 April 2026 · Updated 14 April 2026 · 4 min read
What Is the Poisson Distribution?
The Poisson distribution is a statistical model that predicts the probability of a given number of events occurring in a fixed interval — perfect for predicting the number of goals in a football match, since goals are relatively rare, independent events.
It’s the foundational model behind most bookmaker pricing systems, and understanding it gives you a framework for building your own football prediction model.
Why Poisson Works for Football
Goals in football match several key Poisson assumptions:
- Relatively rare — Most teams score 0-3 goals per match
- Independent — One goal doesn’t directly cause or prevent another (approximately)
- Predictable average rate — Teams have measurable scoring and conceding rates over a season
Step-by-Step: Building a Poisson Model
Step 1: Calculate Attack and Defence Strengths
Using Premier League season data:
League averages (example figures):
- Average home goals per match: 1.52
- Average away goals per match: 1.18
For each team, calculate:
- Home Attack Strength = Team’s home goals scored ÷ League average home goals
- Home Defence Strength = Team’s home goals conceded ÷ League average away goals
- Away Attack Strength = Team’s away goals scored ÷ League average away goals
- Away Defence Strength = Team’s away goals conceded ÷ League average home goals
Step 2: Predict Expected Goals for a Specific Match
Example: Arsenal (home) vs Aston Villa (away)
Home expected goals = Arsenal Home Attack × Villa Away Defence × League Avg Home Goals
Away expected goals = Villa Away Attack × Arsenal Home Defence × League Avg Away Goals
If Arsenal’s home attack strength is 1.45 and Villa’s away defence strength is 1.10:
Arsenal xG = 1.45 × 1.10 × 1.52 = 2.42
If Villa’s away attack strength is 0.95 and Arsenal’s home defence strength is 0.75:
Villa xG = 0.95 × 0.75 × 1.18 = 0.84
Step 3: Calculate Scoreline Probabilities
Using the Poisson formula:
P(x) = (λ^x × e^(-λ)) ÷ x!
Where λ is the expected goals and x is the number of goals.
Arsenal goal probabilities (λ = 2.42):
| Goals | Probability |
|---|---|
| 0 | 8.9% |
| 1 | 21.5% |
| 2 | 26.0% |
| 3 | 21.0% |
| 4 | 12.7% |
| 5+ | 9.9% |
Villa goal probabilities (λ = 0.84):
| Goals | Probability |
|---|---|
| 0 | 43.2% |
| 1 | 36.3% |
| 2 | 15.2% |
| 3 | 4.3% |
| 4+ | 1.0% |
Step 4: Build the Scoreline Matrix
Multiply the home and away probabilities for each scoreline:
P(Arsenal 2, Villa 0) = P(Arsenal 2) × P(Villa 0) = 26.0% × 43.2% = 11.2%
This gives you a complete probability grid for every possible scoreline.
Step 5: Derive Market Probabilities
From your scoreline matrix, calculate:
- Home win probability = Sum of all scorelines where home > away
- Draw probability = Sum of all scorelines where home = away
- Away win probability = Sum of all scorelines where home < away
- Over 2.5 probability = Sum of all scorelines where total > 2
- BTTS probability = Sum of all scorelines where both > 0
Compare these to bookmaker odds to find value.
Limitations of the Poisson Model
- Assumes independence — In reality, one goal can change the dynamic of a match (game state effects)
- Doesn’t account for specific matchups — Playing styles and tactical setups matter
- Historical data may not reflect current form — Season averages smooth out recent changes
- Doesn’t capture extreme events — Red cards, penalties, defensive collapses
- Under-predicts draws — The basic model tends to underestimate the probability of 0-0 and 1-1 draws
Improving Your Model
- Weight recent matches more heavily (e.g., last 10 matches at 60%, season at 40%)
- Adjust for home advantage more precisely (not all home advantages are equal)
- Incorporate xG data instead of actual goals (more predictive)
- Add a draw inflation factor of 3-5% to correct the known bias
- Update after each matchweek to keep the model current
Practical Application
A Poisson model is a starting point, not the final answer. Use it to:
- Generate your own odds for every match
- Compare against bookmaker prices
- Identify where the biggest discrepancies (potential value) exist
- Focus your detailed analysis on those specific matches
This is exactly how many professional betting syndicates operate — the model flags opportunities, and human analysts provide the final decision.
This is an educational guide. No model guarantees betting profits. Always bet responsibly.
18+ only. BeGambleAware.org
Continue Reading
Football Stats for Betting: Which Statistics Actually Matter?
Discover which football statistics actually help with betting. From xG to PPDA, learn the key metrics that predict match outcomes and find betting value.
Premier League xG Table 2025/26: Expected Goals Rankings
Live Premier League expected goals (xG) table for the 2025/26 season. See which teams are overperforming and underperforming based on xG data.
What Is Expected Goals (xG)? Football Analytics Explained
A beginner-friendly explanation of expected goals (xG) in football. Learn what xG means, how it's calculated, and why it's the most important modern football stat.