Belgium vs Iran
Group G · Sunday, June 21, 2026, 19:00 UTC · Los Angeles (Inglewood)
The model's read
The model makes Belgium strong favourites at 54%, leaving 18% for Iran and 27% for the draw. Goals project around 1.7–1.1 (both teams to score 53%, over 2.5 53%). The biggest single factor is Injuries & suspensions, favouring Iran. Confidence sits at 51/100 with moderate upset risk; the underdog has paths to a result.
Auto-updated every hour as ratings, odds and news change.
Match outcome
oracle-v1.0.0
Belgium win
54.5%
Draw
27.1%
Iran win
18.4%
Expected goals
1.67 – 1.12
Most likely exact score
1-0
top exact 1-0 · 12.3%
Confidence
51/100
Result lean
Belgium narrow edge
Score band
Likely one-goal game
Goal environment
Moderate
If it opens up — high-scoring alternative
2-2 at 5.4% — most likely scoreline with 3+ goals total.
top exact score carries 12.3% probability
Risk & uncertainty
Upset risk
40/100
Data quality
83/100
Uncertainty band
38 – 67%
Most likely scorelines
Top 12 from a Dixon-Coles adjusted Poisson matrix.
Top 12 most likely scorelines. Heat is relative to the most likely outcome. Belgium vs Iran.
Team comparison
Rating profile across the key dimensions.
Ensemble breakdown
Final probabilities blend the technical model with every available source layer. Persisted 2026-07-11 01:01:08 UTC.
Final (blended) home
54.5%
Final draw
27.1%
Final away
18.4%
Source weights applied
- technical38.5%
- learned16.1%
- market19.3%
- crowd7.5%
- external0.0%
- intel8.6%
- fundamental7.0%
- sentimentMomentum3.0%
Source presence
- technical1 signal
- learned1 signal
- market37 signals
- crowd1 signal
- externalnot used
- intel62 signals
- fundamental35 signals
Why this prediction
High-coverage prediction — most signal classes are present and aligned.
Technical
74/100
Fundamentals
82/100
Betting
82/100
Prediction market
77/100
Sentiment
0/100
Top positive drivers
- · crowd signal favours home (34.8 pp above uniform)
- · market signal favours home (33.9 pp above uniform)
- · technical signal favours home (19.1 pp above uniform)
- · learned signal favours home (16.6 pp above uniform)
Top risk drivers
- No standout risks.
Missing: external prediction sites
Learned model
The learned model contributes to the public ensemble blend, alongside ratings, market signals and news intelligence. Performance is validated against an Elo baseline.
Belgium win
49.9%
Draw
37.8%
Iran win
12.3%
Total goals (λ)
2.50
Model confidence
30/100
Logistic regression model — version wdl-v1, trained on 19k historical internationals.
Article intelligence
Recent news, injury alerts and sentiment extracted by the NLP layer.
Belgium
50 articles / 14d
Injury alerts
· Belgium (out — out for several weeks)
· Amadou Onana (out — ruled out)
· World Cup (out — sidelined)
Iran
12 articles / 14d
Injury alerts
· Iran (out — ruled out)
· Shoja Khalilzadeh's (out — ruled out)
Goals markets
BTTS yes
52.6%
Over 1.5
74.6%
Over 2.5
52.8%
Over 3.5
30.6%
Under 2.5
47.2%
Market intelligence
Model probability
54.5%
Market probability
67.2%
Gap: -14.8 percentage points.
Market confidence: 96/100 across 37 bookmakers.
Odds shown for analysis only — CupCastLab does not place bets, list bookmakers, or recommend wagers.
Explanation
Why the model sees the match this way.
Model prediction: Belgium 52% to win, 28% for Iran, 20% draw. Expected goals: Belgium 1.67 vs Iran 1.12. Most likely exact score is 1-0 at 12.3%, but no exact score is dominant — the score cluster is 1-0 / 2-1 / 1-1. Goal environment moderate (BTTS 53%, Over 2.5 53%). Likely one-goal game. Primary factor: injuries & suspensions difference is the dominant driver. Confidence: 50/100. Moderate confidence; outcome is meaningfully uncertain.
Factor contributions
Risk factors
- Belgium carry injury concerns into the match.
Uncertainty factors
- No major uncertainty flags.
Tactical read
No standout tactical mismatch — the stylistic edge is modest.
Upset risk
Moderate upset risk — the underdog has identifiable paths to a result.
Market divergence
Notable gap: the market is markedly more bullish than the model. This often reflects public bias, late news, or sharp money the model has not yet incorporated.
These predictions are probabilistic and not guarantees. Not betting advice.