| Outcome | Probability |
|---|---|
| Qatar | 9.4% |
| Draw | 21.4% |
| Switzerland | 69.3% |
| Line | Over | Under |
|---|---|---|
| 0.5 | 92.1% | 7.9% |
| 1.5 | 76.3% | 23.7% |
| 2.5 | 50.6% | 49.4% |
| 3.5 | 28.6% | 71.4% |
| 4.5 | 13.7% | 86.3% |
Poisson total-goals expectation Σλ = 2.7 (Over 2.5 50.6% · Under 2.5 49.4%).
BTTS Yes 42.6% · No 57.4% — neither side dominates the BTTS split.
| Score | Probability |
|---|---|
| 0-2 | 14.2% |
| 0-1 | 13.8% |
| 0-3 | 9.7% |
| 1-2 | 9.1% |
| 1-1 | 8.9% |
Top Poisson cell: 0-2 at 14.2% (exact-score variance remains high).
1X2 from Elo-adjusted λ (0.64 / 2.06): home 9.4%, draw 21.4%, away 69.3%. Source: Dixon–Coles Poisson grid — not bookmaker odds.
Qualitative clarity of the 1X2 split — not a calibrated win-probability confidence interval.
- Lead outcome 69.3% with 47.9 pp over second place.
How are win probabilities calculated?
Home and away expected goals (λ) are derived from Elo ratings and tournament parameters, then fed into a Dixon–Coles Poisson grid to produce 1X2, goal-line, and scoreline probabilities shown on this page.
Is this page betting advice?
No. OddsGPT displays model probabilities for informational purposes only. We do not recommend wagers or stake sizes on this page.
What does xG / λ mean here?
λ is the model’s pre-match expected goals for each team before variance is simulated. It is an input to the Poisson matrix, not a post-match expected-goals stat.
Why are exact score probabilities low?
Even the most likely scoreline typically sits below 15% because many score combinations share the probability mass — that is normal for Poisson models.
