WC2026 Model Tracker
Bivariate-Poisson predictions vs actual results · a live, honest experiment
Iran vs New Zealand UPCOMING
FIFA World Cup 2026, Group G, Matchday 1 · SoFi Stadium, Inglewood CA (indoor, sea level, neutral site)
Kickoff Jun 16, 2026 · 01:00 UTC
Kickoff Jun 16, 2026 · 01:00 UTC
Model vs market
| Market | Model | Book odds | Book p* | EV |
|---|---|---|---|---|
| Iran win | 53.5% | 1.75 | 53.1% | -6.3% |
| Draw | 23.4% | 3.40 | 27.3% | -20.3% |
| New Zealand win | 23.0% | 4.75 | 19.6% | +9.3% |
| Over 2.5 goals | 58.5% | 2.38 | 39.8% | +38.9% |
| Under 2.5 goals | 41.5% | 1.57 | 60.2% | -34.8% |
| Both teams score: yes | 59.7% | 2.20 | 42.3% | +31.3% |
| Both teams score: no | 40.3% | 1.62 | 57.7% | -34.8% |
| Over 8.5 corners (exp 9.49) | 60.7% | 2.10 | 44.2% | +27.4% |
| Under 8.5 corners | 39.3% | 1.67 | 55.8% | -34.5% |
| Over 3.5 cards (exp 3.91) | 54.9% | 1.67 | 55.8% | -8.5% |
| Under 3.5 cards | 45.1% | 2.10 | 44.2% | -5.3% |
| Any red card | 13.1% | — | — | — |
Model-only estimates (no straight market equivalent): expected goals Iran 1.88 – 1.19 New Zealand · fair Asian handicap Iran -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 — Iran vs New Zealand (FIFA World Cup 2026, Group G, Matchday 1)
Venue: SoFi Stadium, Inglewood CA (indoor, sea level, neutral site)
1. DATA. Last-10 internationals (Sofascore interception scraper).
Iran: record DLLWWLLWWW, 2.40 GF / 1.60 GA,
xG 0.51 / xGA 0.72.
New Zealand: record LLLDLLLWLL, 0.70 GF / 1.80 GA,
xG 0.96 / xGA 1.48.
Opponent-strength tiers (configs.MASTER_TIERS) applied per match.
2. RATINGS. Iran atk 1.29 def 1.14; New Zealand atk 0.76 def 1.21.
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 Iran 1.88, New Zealand 1.19
after 85% model weight + shrink to 1.3 baseline.
Injuries: {'Iran': ['No major absences reported'], 'New Zealand': ['No major absences reported']}.
Context: Group G. Iran favoured and well-organised; New Zealand a physical OFC side. Caveat: NZ's recent slate is mostly OFC minnows, so their rating leans on goals and is lower-confidence.
3. MODEL. Bivariate Poisson (lambda3 0.12). 1X2 Iran 53.5% /
draw 23.4% / New Zealand 23.0%. Winner Iran; score 1-1; fair AH Iran -0.75.
Over 2.5 58.5%; BTTS 59.7%.
4. EVENTS. Referee César Ramos (Mexico): 4.18 Y + 0.15 R/match
blended (weight 0.45) with WC base 3.5742 -> expected
cards 3.91, P(over 3.5) 54.9%; P(any red)
13.1%. Corners expected 9.5 (scale
1.0068) -> P(over 8.5) 60.7%.
5. LIMITATIONS. 10-match samples; hand-set tiers/priors; Poisson corners; lineups
probable not confirmed; card base rate calibrated on only 4 WC matches.