Montreal beat Dallas 4-3 in overtime despite losing the expected goals battle 4.65–3.70. This was a heist: the Stars controlled 56% of the expected goals at 5v5 and still went home empty.

Dallas dominated the chance-generation game. The Stars generated 0.088 xG per shot versus Montreal's 0.065—higher-quality scoring chances throughout the night. High-danger looks went 10–3 in Dallas's favor. Yet Montreal's goaltending, particularly S. Montembeault's performance between the pipes, flipped the script.

S. Montembeault finished with a 0.889 save percentage on 24 saves—1.65 goals better than expected. He was especially brilliant on high-danger chances, posting a 90% save percentage on 10 such attempts. J. Oettinger, meanwhile, posted a 0.857 save percentage for Dallas, though the real issue wasn't his performance: the Stars underperformed their expected goals by 1.65 goals, wasting a dominant offensive showcase.

Both teams found rhythm on the rush. Dallas scored twice on 30 rush chances; Montreal answered with three goals on 29 attempts. The difference was conversion when it mattered most.

Lane Hutson's snap shot from the high slot at 03:40 of overtime beat Oettinger. It was the punctuation on a night that validated the oldest axiom in hockey: goaltending and depth scoring can overcome pure firepower.

Dallas enters as a modest favorite—the model projects a 2.4–2.6 goal game with the Stars at 55% win probability. This is about as close to a pick'em as you'll find.

The underlying play belongs to Dallas. The Stars own 51.0% of the 5v5 expected goals against Montreal's 48.0%, a margin that should translate to sustained pressure. Montreal, meanwhile, holds a slight edge in shot volume: 49.0% Corsi against Dallas's 47.0%. That's the fundamental tension—Dallas generates more dangerous chances; Montreal controls the puck more often. One favors Dallas's forwards, the other Montreal's defensemen, and neither is dispositive on its own.

Both teams are running hot and unsustainably so. Dallas sits at 1.024 PDO, Montreal at 1.018—both inflated by shooting luck and goaltending performance that should normalize downward. Expect regression from both.

Cole Caufield is Montreal's engine: 33 goals on 0.97 expected goals per 60 minutes, shooting at 20.0%. That rate performs exactly as five-on-five projections suggest, which means he's neither collapsing nor due for a breakout—just consistent. Oliver Kapanen sits lower in volume (19 goals, 0.83 xG/60) but owns a 17.8% shooting rate, a threshold that typically declines toward league average.

Dallas counters with Jason Robertson (25 goals, 0.96 xG/60) and Michael Bunting (8 goals, 0.83 xG/60). Robertson's expected-goal rate mirrors Caufield's almost exactly. Bunting is simply underproductive at 7.6% shooting—positive regression is coming.

Both Montreal's Mike Matheson and Dallas's Tyler Myers represent defensive liabilities, though the granular data is thin: both teams show 0% high-danger and 0% rebound goals, which suggests sample-size noise in chance categorization rather than structural weakness.

J. Fowler lands on a back-to-back, worth monitoring for either a backup appearance or diminished performance from fatigue.

This is genuinely even. Dallas's expected-goal edge might tip them toward victory, but Montreal's puck possession and Caufield's reliability make an upset entirely plausible. Lean Dallas, but with low conviction.