FINAL: TOR 2 — NYI 1. Our Monte Carlo simulation projected TOR 2.75 - NYI 3.1 (NYI at 51.6% win probability).
TOR
2.75
Projected Score
VS
NYI
3.1
Projected Score
Win Probability
TORNYI
N/A
Spread
N/A
Total Line
10,000
Simulations
NYITOR W4
Calibrated accuracy at this confidence: 51.9% (1,180 games)
Projected Goals Range 10th – 90th percentile
NYI
2.03.14.2
TOR
1.72.83.8
Projected
TOR 2.75 — NYI 3.1
Actual
TOR 2 — NYI 1
Pick Results
NYI MLmlLOSS-0.50u
Goalie Matchup
Ilya Sorokin
29-242.68 GAA90.6% SV
Sergei Bobrovsky
27-233.06 GAA87.5% SV
Special Teams
Power Play
Penalty Kill
90% Confidence: 44.2% – 45.8% home win probability
AI Intelligence Analysis
NEUTRAL -1
Model identifies a real elite-goalie + B2B-fatigue edge favoring NYI (51.57% win prob vs 47.6% market implied), but away underdog ML is our historically POISON category (15W-26L, -16.5u all-time), so the category weakness overrides the game-specific edge.
Key Factors
- Goalie matchup: Sorokin.9059 SV% (rested) vs Stolarz.8927 SV% (B2B fatigued) — 1.32pp gap is significant in a sub-3-goal-per-team environment
- B2B fatigue: TOR rest_days=1, fatigue_factor 0.8836 after a loss to MTL last night — estimated -0.15 to -0.25 goal swing not fully priced by market
- Model-market gap: 51.57% (model) vs 47.6% (market implied) — NYI is underpriced by ~4pp relative to our model's read
- Spread value: model +0.343 vs market -1.5 = ~1.84 goals of edge in NYI's favor, among the largest gaps in recent data
- Category risk: away underdog ML sits at 15W-26L all-time (-16.5u), the single worst bet type/side combo we track
Risk Factors
- Away underdog ML is historically POISON (15W-26L, -16.5u all-time,
- Spread/puckline is F-grade this season (a supportive track record, -3.1u) — not a safe alternative despite the 1.84pt line value
- TOR possesses game-breaking talent (Matthews.359 G/gm, Nylander.777 G/gm in small sample) capable of overcoming goalie/fatigue disadvantages in a single game
Edge Analysis
Moneyline
NYI 51.6%
--
How this prediction was generated: This page shows output from the Olympus Bets NHL Hockey Monte Carlo engine. Each game is simulated 10,000 times using real-time team data, injury reports, and current odds. Probabilities are calibrated using Bayesian methods and sized via the Kelly Criterion. Probabilities are calibrated using Bayesian methods and sized via the Kelly Criterion. Full methodology →
Permanent record: this game page is retired 90 days after the event. The full graded NHL 2026 record — every pick, win rate, units and ROI — lives on the NHL 2026 season report card.