Switzerland
Quarter-finals · 7/12/2026, 1:00:00 AM
1-0
alt 1-1 · xG 1.49–0.73
5conf
actual 3-1 ✓ correct
Score-probability matrix
0
1
2
3
4
5
6
0
11
7
1
16
12
2
12
9
3
6
4
5
6
↓ Argentina goalsSwitzerland goals →
Win / draw / loss
W 55%D 27%L 18%
Market
57% / 27% / 17%
Argentina favoured (55% to win); the model's most likely scoreline is 1-0.
Confidence 5/10. Even the favoured outcome (55%) fails about 45% of the time; a single match is high variance.
How we got to 1-0
- 1Model fitTeam strengths fit from 15,899 matches (Dixon-Coles, recency-weighted, low-score correction ρ=-0.044). Neutral venue — no home advantage.
- 2Team strengths (log scale, vs average)Argentina: attack +1.48, defence +1.80Switzerland: attack +1.28, defence +1.20Higher attack = scores more; higher defence = concedes fewer.
- 3Expected goalsArgentina: exp(base +0.15 + att +1.48 − opp.def +1.20) = 1.54Switzerland: exp(base +0.15 + att +1.28 − opp.def +1.80) = 0.69“base” is the league scoring level the model fitted.
- 4Agent adjustmentsλ 1.54/0.69 → 1.49/0.73
- Tactical away attack ×1.05 — No availability or stakes information beyond what the baseline captures; both sides in strong form. Slight nudge to Switzerland's attack and defence reflects their compact, well-organised tactical setup that tends to overperform their xG baseline against possession-dominant opponents at neutral venues, while Argentina's factors stay neutral.
- Tactical away defence ×1.03 — No availability or stakes information beyond what the baseline captures; both sides in strong form. Slight nudge to Switzerland's attack and defence reflects their compact, well-organised tactical setup that tends to overperform their xG baseline against possession-dominant opponents at neutral venues, while Argentina's factors stay neutral.
- 5ScorelineA Poisson model on those expected goals (with ρ) gives the score-probability matrix above; its most likely cell is 1-0 at 16%. Win / draw / loss come from summing the matrix.
- 6Confidence5/10 — a sharper peak and a clearer favourite raise it; large agent moves lower it (model agreement 93%).