FINAL: NJD 3 — UTA 5. Our Monte Carlo simulation projected NJD 2.68 - UTA 3.28 (UTA at 51.6% win probability).
NJD
2.68
Projected Score
VS
UTA
3.28
Projected Score
Win Probability
NJDUTA
N/A
Spread
N/A
Total Line
10,000
Simulations
UTA W5NJD L5
Calibrated accuracy at this confidence: 51.6% (1,222 games)
Projected Goals Range 10th – 90th percentile
UTA
2.23.34.4
NJD
1.62.73.8
Projected
NJD 2.68 — UTA 3.28
Actual
NJD 3 — UTA 5
Pick Results
UTA MLmlWIN+0.57u
Goalie Matchup
Karel Vejmelka
38-201.52 GAA91.7% SV
Jake Allen
17-171.89 GAA92.0% SV
Special Teams
Power Play
Penalty Kill
90% Confidence: 38.9% – 40.5% home win probability
AI Intelligence Analysis
NEUTRAL
UTA's superior recent form (+2.33 L5 differential) and better defensive archetype rating partially offset NJD's home ice and Allen's edge in net, leaving a genuinely mixed signal that doesn't clear our away-dog bar for action.
Key Factors
- Goalie matchup is close to a wash: Allen.920 SV%/1.89 GAA (NJD, elite tier) vs Vejmelka.917 SV%/1.52 GAA (UTA, starter tier) — minimal edge either way.
- UTA's archetype (balanced_elite, def structure 1.062) outranks NJD's balanced profile (def structure 0.994), suggesting UTA is the more efficient defensive team early this season.
- L5 goal differentials tell opposite stories: UTA +2.33 (4.0 GF/1.67 GA) vs NJD -2.5 (1.5 GF/4.0 GA) — UTA has been markedly better recently.
- Model implies only 48.4% win prob for NJD despite a -125 market price (55.6% implied) — a 7.2-point gap that isn't cleanly explained by goalie or injury data.
- UTA carries 3-in-4 fatigue (factor 0.90), a modest offsetting concern for their superior form.
Risk Factors
- Away underdog historical support carries a historical 37.2-a supportive track record penalty that even solid UTA form doesn't clearly overcome.
- NJD's home ice (+0.25 goals) plus Allen's confirmed elite tier are real, if modest, factors in their favor that could be enough on a given night.
Edge Analysis
Moneyline
UTA 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.