WC2026 Model Tracker
Bivariate-Poisson predictions vs actual results · a live, honest experiment
🇧🇪 Belgium vs Iran WINNER MISS
FIFA World Cup 2026, Group G, Matchday 3 · Neutral site (USA)
Kickoff Jun 21, 2026 · 19:00 UTC
Kickoff Jun 21, 2026 · 19:00 UTC
Final · Belgium 0–0 Iran
Goals: —
Actual stats: xG 1.79 – 0.62 · 6 corners · 3 cards (1 red)
Prediction vs result 1/8 markets hit
| Market | Model called | Actual | Outcome |
|---|---|---|---|
| 1X2 | Belgium | Draw (0-0) | MISS |
| Score | 1-1 | 0-0 | MISS |
| O/U 2.5 | Over 2.5 (61.5%) | 0 goals | MISS |
| BTTS | Yes (61.4%) | No | MISS |
| Corners | Over 8.5 (62.5%) | 6 corners | MISS |
| Cards | Under 3.5 (50.7%) | 3 cards | HIT |
| Red card | No (84.8%) | Yes | MISS |
| AH | Belgium -0.75 | Loss | MISS |
Model vs market
| Market | Model | Book odds | Book p* | EV |
|---|---|---|---|---|
| Belgium win | 55.7% | 1.38 | 68.8% | -23.1% |
| Draw | 22.4% | 4.90 | 19.4% | +9.8% |
| Iran win | 21.9% | 8.00 | 11.9% | +75.0% |
| Over 2.5 goals | 61.5% | 1.80 | 52.6% | +10.7% |
| Under 2.5 goals | 38.5% | 2.00 | 47.4% | -23.0% |
| Both teams score: yes | 61.4% | 2.05 | 45.3% | +25.8% |
| Both teams score: no | 38.6% | 1.70 | 54.7% | -34.4% |
| Over 8.5 corners (exp 9.64) | 62.5% | — | — | — |
| Under 8.5 corners | 37.5% | — | — | — |
| Over 3.5 cards (exp 3.64) | 49.3% | — | — | — |
| Under 3.5 cards | 50.7% | — | — | — |
| Any red card | 15.2% | — | — | — |
Model-only estimates (no straight market equivalent): expected goals Belgium 2.00 – 1.22 Iran · fair Asian handicap Belgium -0.75
In the table above, Model = our probability and Book odds / Book p* = the bookmaker's (p* has the margin removed). EV = value per unit staked at the model's probability (positive = model sees value). “—” = market not in the captured feed.
Model reasoning (chain of thought)
PREDICTION RATIONALE — Belgium vs Iran (FIFA World Cup 2026, Group G, Matchday 3)
Venue: Neutral site (USA)
1. DATA. Last-10 internationals (Sofascore interception scraper).
Belgium: record WDWDWWDWWD, 3.20 GF / 0.70 GA,
xG 2.38 / xGA 0.59.
Iran: record LLWWLLWWWD, 2.40 GF / 1.60 GA,
xG 0.84 / xGA 0.89.
Opponent-strength tiers (configs.MASTER_TIERS) applied per match.
2. RATINGS. Belgium atk 1.48 def 0.70; Iran atk 1.27 def 1.13.
NEUTRAL VENUE: home boost / away penalty set to 1.0/1.0 (both teams
travel to a host-nation site; no host advantage applies — a key difference
from the Korea/Canada/USA host-nation games). lambda Belgium 2.00, Iran 1.22
after 85% model weight + shrink to 1.3 baseline.
Injuries: {'Belgium': ['No major absences reported'], 'Iran': ['No major absences reported']}.
Context: Group G MD3. Belgium favoured; Iran a real AFC side. Neutral venue. MD3 rotation risk if group is decided.
3. MODEL. Bivariate Poisson (lambda3 0.12). 1X2 Belgium 55.7% /
draw 22.4% / Iran 21.9%. Winner Belgium; score 1-1; fair AH Belgium -0.75.
Over 2.5 61.5%; BTTS 61.4%.
4. EVENTS. Referee Darío Herrera (ARG): 5.2 Y + 0.2 R/match
blended (weight 0.29) with WC base 2.9307 -> expected
cards 3.64, P(over 3.5) 49.3%; P(any red)
15.2%. Corners expected 9.6 (scale
0.9785) -> P(over 8.5) 62.5%.
5. LIMITATIONS. 10-match samples; hand-set tiers/priors; Poisson corners; lineups
probable not confirmed; card base rate calibrated on only 4 WC matches.