WC2026 Model Tracker
Bivariate-Poisson predictions vs actual results · a live, honest experiment
Tunisia vs Japan WINNER HIT
FIFA World Cup 2026, Group F, Matchday 3 · Neutral site (USA)
Kickoff Jun 21, 2026 · 04:00 UTC
Kickoff Jun 21, 2026 · 04:00 UTC
Final · Tunisia 0–4 Japan
Goals: 4' Daichi Kamada (Japan); 31' Ayase Ueda (Japan); 69' Junya Ito (Japan); 83' Ayase Ueda (Japan)
Actual stats: xG 0.05 – 2.13 · 8 corners · 0 cards (0 red)
Prediction vs result 5/8 markets hit
| Market | Model called | Actual | Outcome |
|---|---|---|---|
| 1X2 | Japan | Japan (0-4) | HIT |
| Score | 0-1 | 0-4 | MISS |
| O/U 2.5 | Over 2.5 (53.5%) | 4 goals | HIT |
| BTTS | Yes (54.5%) | No | MISS |
| Corners | Under 8.5 (52.7%) | 8 corners | HIT |
| Cards | Under 3.5 (55.2%) | 0 cards | HIT |
| Red card | No (87.8%) | No | HIT |
| AH | Tunisia +0.75 | Loss | MISS |
Model vs market
| Market | Model | Book odds | Book p* | EV |
|---|---|---|---|---|
| Tunisia win | 19.8% | 6.50 | 14.5% | +28.6% |
| Draw | 23.9% | 4.00 | 23.6% | -4.4% |
| Japan win | 56.3% | 1.53 | 61.8% | -13.8% |
| Over 2.5 goals | 53.5% | 2.10 | 45.1% | +12.3% |
| Under 2.5 goals | 46.5% | 1.73 | 54.9% | -19.6% |
| Both teams score: yes | 54.5% | 2.10 | 44.2% | +14.5% |
| Both teams score: no | 45.5% | 1.67 | 55.8% | -24.2% |
| Over 8.5 corners (exp 8.47) | 47.3% | 1.73 | 53.7% | -18.4% |
| Under 8.5 corners | 52.7% | 2.00 | 46.3% | +5.5% |
| Over 3.5 cards (exp 3.43) | 44.8% | — | — | — |
| Under 3.5 cards | 55.2% | — | — | — |
| Any red card | 12.2% | — | — | — |
Model-only estimates (no straight market equivalent): expected goals Tunisia 1.02 – 1.83 Japan · fair Asian handicap Tunisia +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 — Tunisia vs Japan (FIFA World Cup 2026, Group F, Matchday 3)
Venue: Neutral site (USA)
1. DATA. Last-10 internationals (Sofascore interception scraper).
Tunisia: record WWLDLWDLLL, 1.30 GF / 2.10 GA,
xG 0.67 / xGA 1.61.
Japan: record DLDWWWWWWD, 1.50 GF / 0.80 GA,
xG 1.34 / xGA 1.38.
Opponent-strength tiers (configs.MASTER_TIERS) applied per match.
2. RATINGS. Tunisia atk 0.91 def 1.39; Japan atk 1.16 def 0.75.
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 Tunisia 1.02, Japan 1.83
after 85% model weight + shrink to 1.3 baseline.
Injuries: {'Tunisia': ['No major absences reported'], 'Japan': ['No major absences reported']}.
Context: Group F MD3. Japan favoured; Tunisia defensive. Neutral venue. MD3 rotation risk if group is decided.
3. MODEL. Bivariate Poisson (lambda3 0.12). 1X2 Tunisia 19.8% /
draw 23.9% / Japan 56.3%. Winner Japan; score 0-1; fair AH Tunisia +0.75.
Over 2.5 53.5%; BTTS 54.5%.
4. EVENTS. Referee TBD: 4.2 Y + 0.13 R/match
blended (weight 0.33) with WC base 2.9923 -> expected
cards 3.43, P(over 3.5) 44.8%; P(any red)
12.2%. Corners expected 8.5 (scale
0.9997) -> P(over 8.5) 47.3%.
5. LIMITATIONS. 10-match samples; hand-set tiers/priors; Poisson corners; lineups
probable not confirmed; card base rate calibrated on only 4 WC matches.