| Score | Likelihood | Prob | Fair odds | |
|---|---|---|---|---|
| 3–1MODE | 7.3% | 13.68 | ||
| 2–1 | 7.2% | 13.89 | ||
| 3–2 | 6.1% | 16.37 | ||
| 2–2 | 6.0% | 16.62 | ||
| 4–1 | 5.6% | 17.96 | ||
| 1–1 | 4.7% | 21.16 | ||
| 4–2 | 4.7% | 21.49 | ||
| 3–0 | 4.4% | 22.87 |
Canada scores shown first. Bivariate Poisson with team xG from recent form — most-likely score ≠ most-likely winner.
Probabilities derived from a Poisson model on expected goals. Edge % shown only when the book posts both sides of a market.
| Combo | Joint prob | Naïve (book) | Corr. | Fair odds |
|---|---|---|---|---|
Home win + BTTS home + btts | 46.3% | 49.7% | -7% | 2.16 |
Home win + Over 2.5 home + total_goals_2.5 | 57.1% | 54.5% | +5% | 1.75 |
Away win + BTTS away + btts | 15.4% | 15.0% | +3% | 6.50 |
Away win + Over 2.5 away + total_goals_2.5 | 16.5% | 16.5% | +0% | 6.05 |
Joint probability comes from the score grid (the truth). Naïve is what a book gets by multiplying legs — they undershoot correlated positive legs like "home + BTTS" and overshoot anti-correlated ones like "draw + over 2.5". Buy positive correlation; avoid negative.
Model probabilities Canada 45.3% · Draw 25.8% · Morocco 28.9% Why Canada is favoured Elo essentially level (1527 vs 1534) — winner depends on context. Recent: Canada 2W-1D-1L · Morocco 3W-1D-0L (last 4). Goal differential per game: Canada +1.5, Morocco +1.0. Form depth Canada: avg 2.3 goals scored, 0.8 conceded per game (1/2 at home, 1/2 on the road). Morocco: avg 2.5 goals scored, 1.5 conceded per game (1/1 at home, 2/3 on the road). Morocco on a 3-match winning streak. Goals expectation Model expects ~3.5 total goals (Canada 1.9, Morocco 1.6) — above the 2.5-goal mid line. But — counter-arguments Elo confidence is low (20%) — ratings are based on a thin sample, so the prediction itself is wider than the headline number suggests. Context Canada has 1 more day of rest (Canada 20d, Morocco 19d). Confidence note Elo data on these teams is thin (20% sample-size confidence); this is a low-conviction match overall — bet only with strong reason from outside the model.