Algorithm Methodology & Transparency

Last updated: 2026-04-19 · Version 14.8 · Contact: footballinteligenceAI@gmail.com

1. Data Sources & League Coverage

Football Intelligence ingests historical and live match data from three primary sources:

Leagues covered: Premier League (England), La Liga (Spain), Bundesliga (Germany), Serie A (Italy), Ligue 1 (France), UEFA Champions League, UEFA Europa League, UEFA Conference League, Eredivisie (Netherlands), Primeira Liga (Portugal), Ekstraklasa (Poland), Scottish Premiership, Belgian Pro League.

2. The Poisson Distribution Model

The Poisson distribution is a discrete probability distribution that expresses the probability of a given number of events (goals) occurring in a fixed interval, given a known average rate (lambda, λ). In football, it was shown by Maher (1982) and Dixon & Coles (1997) that goals in a match closely follow independent Poisson processes for each team.

The core formula is:

By computing P(home_goals = i) and P(away_goals = j) for all i, j from 0–7, we generate a full 8×8 scoreline probability matrix. Summing the appropriate cells gives: win probability (Σ cells where i > j), draw probability (Σ diagonal), and away win probability (Σ cells where j > i).

3. Attack Strength & Defensive Record

For each team playing at home, we compute two relative indices from their season-to-date results:

These indices measure performance relative to the league average. A value of 1.0 means average; above 1.0 means better than average at that role.

Lambda calculation for a domestic match:

League average goals by competition are maintained in a calibration table updated each season. Example values: Premier League home 1.53, away 1.19; Bundesliga home 1.72, away 1.28.

4. Bayesian Regularisation

Raw attack/defense indices are unreliable when a team has played few matches (e.g., early in the season). We apply Bayesian regularisation (shrinkage toward the prior mean of 1.0):

This ensures that teams with fewer than 6 matches do not generate extreme predictions. The sample window is capped at 14 games to maintain recency: only the most recent half of a team's games are used to compute the index (weighted toward recent form).

5. Home Advantage Weighting

Home advantage is intrinsically encoded in the separate home/away league averages used as multipliers. We do not apply a fixed additive home-advantage constant; instead, the multiplicative structure naturally captures the league-specific home advantage through the league_avg_home_goals parameter, which is calibrated from 5+ seasons of historical data per competition.

6. European Competition Adjustments

For UEFA Champions League and Europa League matches, teams come from different domestic leagues with different strength levels. We normalise using the UEFA country coefficient ranking (5-year method):

Each team's domestic lambda is scaled by the ratio of their league's strength factor. European matches use neutral-ground averages (home: 1.40, away: 1.18) as the baseline, reflecting reduced home advantage in two-legged European ties.

7. Scoreline Probability Matrix

Using λ_home and λ_away, we compute P(i, j) for all scorelines i ∈ [0,7] and j ∈ [0,7]. From this 8×8 matrix we derive:

• Home win probability: Σ P(i,j) where i > j
• Draw probability: Σ P(i,j) where i = j
• Away win probability: Σ P(i,j) where j > i
• BTTS (Both Teams To Score): 1 − P(home=0) − P(away=0) + P(0,0)
• Over 2.5 goals: Σ P(i,j) where i + j ≥ 3
• Exact score probabilities: each P(i,j) cell directly

Fair decimal odds are computed as the inverse of the probability: fair_odds = 1 / probability. Unlike bookmaker odds, these contain no overround margin.

8. Value Betting Detection

A value bet exists when a bookmaker's offered decimal odds are higher than the fair odds derived by our model:

Football Intelligence fetches live odds from 40+ sportsbooks via The Odds API and displays the highest available odds for each market. Only bets with a positive value edge are highlighted as opportunities. A value edge of 5%+ is generally considered meaningful for long-term profitability.

9. Kelly Criterion Stake Calculator

The Kelly Criterion (Kelly, 1956) is a mathematical formula for the theoretically optimal fraction of a bankroll to stake on a bet with known edge and odds:

We display both Full Kelly and Half-Kelly (50% of the calculated fraction). We strongly recommend Half-Kelly or lower, because:

1. No model has perfect probability estimation — model error means actual edge is likely lower than calculated.
2. Full Kelly leads to high variance and bankroll drawdowns even with a genuine edge.
3. The Kelly formula assumes independent bets and a long-run perspective.

10. Claude Sonnet AI Layer — SportGPT

SportGPT is a conversational AI assistant powered by Claude Sonnet 4.6 (Anthropic, 2024). It receives the Poisson model output for a specific match as context and can answer natural-language questions about probability breakdowns, form analysis, staking recommendations, and market interpretation.

SportGPT does not independently calculate probabilities — it interprets and explains the outputs of the Poisson model described above. The AI layer adds natural-language reasoning, contextual knowledge (injuries, team news, head-to-head history from its training data), and personalised staking advice based on the user's bankroll and risk profile.

11. Model Limitations & Accuracy

We are committed to transparent communication about what our model cannot do:

• Football is high-variance: Even the most accurate statistical model cannot reliably predict individual match outcomes. A team with a 70% win probability still loses 30% of the time.
• In-game events not included: The model does not account for in-match red cards, tactical substitutions, or weather conditions in real time.
• Injury data: The model uses publicly available data. Undisclosed injuries or late team-news changes are not reflected until the model is retrained.
• Market efficiency: Modern bookmaker odds are very efficient. Genuine value edges of 5%+ are rare and require large samples to realise.
• No guarantee of profit: Past model performance does not predict future results. Value betting requires discipline over hundreds of bets to see statistical advantage materialise.
• Small sample leagues: For leagues with fewer historical seasons in our dataset, predictions carry higher uncertainty and lambda values are more regularised toward league averages.

Football Intelligence displays model accuracy metrics and a settled pick history so users can evaluate actual performance over time. We do not cherry-pick results or hide losses.

12. Legal Disclaimer

Questions about our methodology? Contact us at footballinteligenceAI@gmail.com