Canada vs Bosnia & Herzegovina

WC2026 Model Tracker

Bivariate-Poisson predictions vs actual results · a live, honest experiment

🇨🇦 Canada vs Bosnia & Herzegovina 🇧🇦 WINNER MISS

FIFA World Cup 2026, Group B, Matchday 1 · BMO Field, Toronto, Canada (elevation 83 m)
Kickoff Jun 12, 2026 · 19:00 UTC

Final · Canada 1–1 Bosnia & Herzegovina

Goals: 21' Jovo Lukić (Bosnia & Herzegovina); 78' Cyle Larin (Canada)

Actual stats: xG 1.25 – 0.98 · 13 corners · 5 cards (0 red)

Prediction vs result 3/8 markets hit

Market	Model called	Actual	Outcome
1X2	Canada	Draw (1-1)	MISS
Score	1-0	1-1	MISS
O/U 2.5	Under 2.5 (63.3%)	2 goals	HIT
BTTS	No (55.8%)	Yes	MISS
Corners	Under 8.5 (54.8%)	13 corners	MISS
Cards	Over 3.5 (59.0%)	5 cards	HIT
Red card	No (84.8%)	No	HIT
AH	Canada -0.25	Half Loss	MISS

Model vs market

Market	Model	Book odds	Book p*	EV
Canada win	43.6%	1.80	52.6%	-21.5%
Draw	30.9%	3.60	26.3%	+11.3%
Bosnia & Herzegovina win	25.5%	4.50	21.1%	+14.8%
Over 2.5 goals	36.7%	2.30	41.3%	-15.7%
Under 2.5 goals	63.3%	1.62	58.7%	+2.3%
Both teams score: yes	44.2%	2.05	45.3%	-9.3%
Both teams score: no	55.8%	1.70	54.7%	-5.2%
Over 8.5 corners (exp 8.32)	45.2%	1.80	51.5%	-18.7%
Under 8.5 corners	54.8%	1.91	48.5%	+4.6%
Over 3.5 cards (exp 4.12)	59.0%	—	—	—
Under 3.5 cards	41.0%	—	—	—
Any red card	15.2%	—	—	—

Model-only estimates (no straight market equivalent): expected goals Canada 1.25 – 0.91 Bosnia & Herzegovina · fair Asian handicap Canada -0.25

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 — Canada vs Bosnia & Herzegovina, WC2026 Group B (BMO Field, 2026-06-12 19:00 UTC)

1. DATA. Last 10 internationals via Sofascore interception scraper.
   Canada (record WLDDWWDDWD): 0.90 GF / 0.40 GA, xG 1.22 / xGA 0.82, possession 49.3%.
   Bosnia (record WLDWWDWWDD): 2.80 GF / 1.30 GA, xG 1.98 / xGA 1.28, possession 51.4%.
   Opponent-strength tiers applied per match as in the KOR-CZE model.

2. RATINGS. Canada atk 0.95 def 0.65; Bosnia atk 1.84 def 1.55.
   Weights loaded from model_weights.json (post KOR-CZE conservative update):
   home boost 1.1059 (was 1.06 pre-update) applied to Canada — true home
   nation, sold-out BMO Field, no travel; away-context penalty 0.917
   applied to Bosnia — transatlantic relocation, flat warm-ups (0-0 N.Macedonia,
   1-1 Panama), hostile crowd. No altitude factor (Toronto, 83 m).
   lambda(CAN) 1.25, lambda(BIH) 0.91 after 85% model weight /
   shrink to 1.30 baseline.
   Key absences priced qualitatively: Davies (LB, out) weakens Canada's left
   side but Laryea is an adequate deputy; Bombito out thins CB depth. Dzeko
   (40) starts for Bosnia — focal point but limited pressing.

3. MODEL. Bivariate Poisson, lambda3 0.12, matrix to 8 goals.
   1X2: CAN 43.6% / draw 30.9% / BIH 25.5%.
   Predicted winner: Canada; most likely score 1-0.
   Over 2.5: 36.7%; BTTS yes: 44.2%.

4. EVENTS. Referee Facundo Tello: 4.59 Y + 0.20 R per match career (ARG league
   sample runs hot — weight 0.48 on career vs WC base 3.5, reduced
   from 0.50 after the Amin Omar miss) -> expected cards 4.12,
   P(over 3.5) 59.0%; red-card Poisson rate 0.165 -> P(any red) 15.2%.
   Corners: expected 8.3 (scale 0.9978) -> P(over 8.5) 45.2%.

5. WEATHER. 26.6 C, 42% RH, 0% rain, wind 21.3 km/h —
   warm and dry; slight tempo tax late, no directional edge.

6. LIMITATIONS. Same as KOR-CZE model (10-match samples, hand-set tiers,
   Poisson corners). Bosnia's qualifying sample is UEFA-heavy vs Canada's
   friendly-heavy sample; tiers mitigate but don't eliminate the gap. The
   learned away-context penalty originated as an altitude-fatigue term; its
   reuse here as a generic away penalty is a modeling assumption.