The Canucks won a game they had no business winning. Vancouver took a 4–3 victory from Washington despite losing the expected-goals battle 5.40–3.06—a 64–36 split so lopsided it made the final score look like an arithmetic error.

The Capitals' dominance was comprehensive. They won 31 shot attempts to 24, generated nine high-danger chances to three, and held 56% of shot volume. Their power play was surgical: two goals on 23 shots (1.90 xG). On paper, Washington controlled this game completely.

T. Demko broke the script. The Vancouver goalie made 28 saves on 31 shots (0.903 SV%), finishing +2.40 against expected goals—meaning he stopped 2.4 more goals than the Capitals should have scored. C. Lindgren, Washington's starter, cut the other direction: 20 saves on 24 shots (0.833 SV%), sitting –0.94, a performance that handed the Canucks the margin.

Vancouver's winning came from two edges. The Canucks exploited the rush ruthlessly, scoring two goals from 22 chances. The Capitals created more rebound opportunities (eight) than Vancouver (seven), but converted just one to Vancouver's two. The goaltending mismatch decided it.

John Carlson's snap shot from the point at 17:46 of the third was Washington's best chance, though it proved immaterial.

The Three Stars:

• T. Myers (VAN, #57): 1G, 1A in 26:12
• C. Garland (VAN, #8): 2A in 22:20
• T. Wilson (WSH, #43): 2A in 20:57

The model projects Washington at 57% to win, targeting a 2.4–2.9 goal scoresheet. That edge rests on underlying play: Washington dominates shot quality with 50.0% 5v5 expected goals against Vancouver's 44.0%. Corsi% tells the same story (49.0% for Washington, 46.0% for Vancouver).

But Washington is running unsustainably hot. Its 1.018 PDO—the all-in measure of puck luck—will regress. Vancouver, sitting at 0.974, is underperforming its chances and faces positive regression in the other direction. The model's edge, grounded in shot quality, narrows once luck normalizes.

Goaltending adds uncertainty. K. Lankinen faces a back-to-back. Whether Washington deploys a backup or runs its starter on fumes is unknowable until puck drop, but fatigue becomes a variable the model cannot forecast.

Filip Chytil is Vancouver's most dangerous weapon. At 1.03 xG/60, he generates at an elite rate, converted 3 goals on 15.0% shooting—room for regression, but a threat tonight. Linus Karlsson (0.95 xG/60, 13G 16A, 14.1% shooting) spreads that load, steady and efficient. Victor Mancini profiles as Vancouver's defensive vulnerability.

Washington counters with Alex Ovechkin at 0.97 xG/60, posting 19 goals and 17 assists on 13.8% shooting. Not Chytil-hot, but elite volume and deployment move the needle. Anthony Beauvillier (0.89 xG/60, 14G 9A, 10.4% shooting) adds secondary punch. Declan Chisholm represents Washington's defensive liability.

Neither team has generated high-danger goals or rebounds at meaningful rates—both register 0% for those buckets—so the advantage concentrates on baseline shot generation and which goaltender stays fresher.