DAN AI SPORTS
/WORLD CUP 2026
SIGN UP FREE
WORLD CUP 2026PREMIER LEAGUECHAMPIONS LEAGUELA LIGABUNDESLIGASERIE AEFLFA CUPPICKLEBALL
CRICKETSOON
NFLSOON
← WORLD CUP HUB/METHODOLOGY

How the Model Works

Every probability on this site comes from the same source: a Dixon-Coles calibrated Poisson model. Here is exactly how it works, what the numbers mean, and where the model is weakest.

1. Poisson Goal Scoring

Goals in football follow a Poisson distribution — they arrive at a roughly constant rate, independently of each other. If we expect a team to score an average of λ goals, the probability of scoring exactly k goals is:

P(X = k) = (λᵏ × e⁻λ) / k!

We calculate a separate λ for each team (home xG, away xG) and convolve the two distributions to get a full probability grid of every possible scoreline from 0-0 to 5-5. From that grid we can read off 1X2 probabilities, over/under lines, BTTS, Asian handicap — every market you see.

2. Dixon-Coles Correction

The standard Poisson model over-predicts 0-0 and 1-0 scorelines relative to real football data. Dixon & Coles (1997) introduced a correction factor ρ that adjusts the joint probability of low-scoring results:

τ(0,0) = 1 − ρ · λ_H · λ_A
τ(1,0) = 1 + ρ · λ_A
τ(0,1) = 1 + ρ · λ_H
τ(1,1) = 1 − ρ

We use ρ = −0.11, calibrated to World Cup historical data. This slightly reduces the probability of 0-0 and 1-1 draws and increases 2-0, 2-1 etc, which better reflects real tournament football.

INFO
Reference: Dixon, M.J. & Coles, S.G. (1997). “Modelling Association Football Scores and Inefficiencies in the Football Betting Market.”Journal of the Royal Statistical Society: Series C, 46(2), 265-280.

3. Attack & Defense Strength Ratings

Each team gets an attack rating and a defense rating derived from their goals scored and conceded per game relative to the tournament average. A rating of 1.0 is exactly average.

attack = (GF per game) / (tournament avg goals per game)
defense = (GA per game) / (tournament avg goals per game)
λ_home = attack_H × defense_A × avg_goals
λ_away = attack_A × defense_H × avg_goals

A high attack rating means a team scores more than average. A high defense rating means they concede more than average (so a low defense rating = good defense). The expected goals for each team is then their attack strength times the opponent’s defensive weakness times the baseline tournament scoring rate.

WARN
Early-tournament caveat: With only 1-3 games played, strength estimates are noisy. A team that won 3-0 in game 1 will look like a +2.5 attack team when they may just have had a weak opponent. Treat model probabilities in the first week with appropriate skepticism. Accuracy improves as sample size grows.
LIVE TEAM STRENGTH RATINGS · UPDATED FROM GROUP STAGE
TeamGPGFGAAttack ↑Defense ↓xG Proj (vs avg)
Germany1714.400.634.400.57
Sweden1513.150.633.150.57
United States1412.520.632.520.57
Norway1412.520.632.520.57
England1422.521.262.521.13
Canada2712.200.312.200.28
France1311.890.631.890.57
Argentina1301.890.001.890.00
Austria1311.890.631.890.57
Colombia1311.890.631.890.57
Switzerland2521.570.631.570.57
Australia1201.260.001.260.00
Netherlands1221.261.261.261.13
Japan1221.261.261.261.13
Iran1221.261.261.261.13
New Zealand1221.261.261.261.13
Croatia1241.262.521.262.27
Mexico2300.940.000.940.00
Czechia2230.630.940.630.85
South Korea2220.630.630.630.57
Bosnia-Herzegovina2250.631.570.631.42
Brazil1110.630.630.630.57
Scotland1100.630.000.630.00
Morocco1110.630.630.630.57
Paraguay1140.632.520.632.27
Ivory Coast1100.630.000.630.00
Curaçao1170.634.400.633.96
Tunisia1150.633.150.632.83
Belgium1110.630.630.630.57
Egypt1110.630.630.630.57
Uruguay1110.630.630.630.57
Saudi Arabia1110.630.630.630.57
Senegal1130.631.890.631.70
Iraq1140.632.520.632.27
Jordan1130.631.890.631.70
Portugal1110.630.630.630.57
Uzbekistan1130.631.890.631.70
Congo DR1110.630.630.630.57
Ghana1100.630.000.630.00
South Africa2130.310.940.310.85
Qatar2170.312.200.311.98
Haiti1010.000.630.000.57
Türkiye1020.001.260.001.13
Ecuador1010.000.630.000.57
Spain1000.000.000.000.00
Cape Verde1000.000.000.000.00
Algeria1030.001.890.001.70
Panama1010.000.630.000.57

Attack ↑ = higher is better · Defense ↓ = lower is better · refreshes every 5 min from ESPN

4. Closing Line Value (CLV)

CLV is the gold standard metric for evaluating betting skill. It measures whether your price was better than the closing market price — which, after removing vig, represents the sharpest collective estimate of true probability.

CLV = model probability − closing de-vigged implied probability

De-vig: implied(home) = (1/homeOdds) / (1/homeOdds + 1/drawOdds + 1/awayOdds)

Positive CLV means the model found value the market hadn’t fully priced in at closing. Sustained positive CLV across a large sample is the strongest evidence of genuine edge. Negative CLV means the market was sharper than the model on that outcome.

INFO
CLV data only accumulates from matches where we have both a model prediction AND closing odds. Matches played before odds data was captured will show “no odds data” in the CLV column.

5. Kelly Staking & Bankroll Simulation

The Kelly Criterion tells you the optimal fraction of your bankroll to bet to maximise long-run growth, given your perceived edge:

Kelly fraction = (prob × odds − 1) / (odds − 1)

We use ¼ Kelly (25% of the full Kelly fraction) for all simulations and recommendations. Full Kelly is theoretically optimal but very volatile — most professional bettors use half or quarter Kelly to reduce variance while retaining most of the growth benefit. We also cap any single bet at 5% of bankroll.

WARN
The bankroll simulation starting at $1,000 is a retrospective illustration only. Past model accuracy does not predict future results. This is not financial advice.

6. Where the Model Is Weakest

  • Early group stage: strength ratings based on 1-2 games are very noisy
  • Knockout rounds: single-elimination changes incentives; tournament fatigue matters
  • Injuries & suspensions: the model doesn’t read team news; market prices often do
  • Penalties: all markets exclude knockout penalty shootouts; these are ~50/50 regardless
  • Home advantage: not modelled (no true “home” team at a neutral tournament)
  • Weather & altitude: we display altitude badges but don’t adjust λ values for them
AI-generated analysis for entertainment purposes only. Nothing on this site constitutes financial or betting advice. Always bet within your means. 18+. BeGambleAware.org
AI-generated analysis for entertainment purposes only. All panellists are AI agents. Odds are indicative and may have changed — verify before placing any bet. Nothing on this site constitutes financial or betting advice. Please gamble responsibly. 18+ only. BeGambleAware.org
Free calculators →Model methodology →