VGK fired 27 shots to Seattle's 22, an advantage that usually decides games. Instead, the Kraken's execution and J. Daccord's 0.963 SV% produced a 2–1 overtime win.

The underlying numbers confirm the gap was razor-thin. Expected goals ran 3.42–3.63, a 49–51 split that reads like a coin flip. But Vegas recorded that 3.42 xG and scored once — 2.42 goals shy of expectation. Daccord was the inverse: he saved 2.42 goals above expectation and was perfect on 5 high-danger chances. That disparity, not volume, decided the night.

Vegas couldn't finish. They surrendered 2.42 goals' worth of chances (0-for-5 on high-danger work), while Daccord was 5-for-5 on the same metric. A. Hill (20 saves, 0.909 SV%) wasn't the bottleneck. The Kraken engineered their victory through relentless net-front pursuit: three rebounds created, one converted. Jared McCann's wrist shot from the doorstep at 4:56 of overtime was the game-winner.

The model favors Vegas at 55% win probability, projecting a 2.5–2.2 Golden Knights victory. This is pick'em territory, and the thin edge exists for one reason: Vegas controls play but runs unsustainably cold.

The numbers heavily favor Vegas. At 5v5, the Golden Knights generate 54.0% of expected goals to Seattle's 46.0% and maintain 52.0% Corsi, signaling territorial dominance. Yet over their last 10 games, Vegas sits at -7 goal differential. The culprit: a PDO of 0.983—below league average and indicating both shooting and goaltending luck have abandoned them. That ratio is unsustainable, which embeds positive regression into the model's call.

Seattle's 1.011 PDO flips the script. The Kraken outperform underlying metrics through elite shooting and goaltending performance. When regression strikes—and it will—Vegas' play advantage converts to goals.

The goaltending matchup adds noise. Both J. Daccord and A. Schmid are on back-to-backs, creating uncertainty around starters or fatigue. A fresh Vegas netminder against a backup widens the model's edge.

Vegas' offensive load concentrates on two weapons. Brandon Saad (0.91 xG/60, 3G, 6A, 1 HD goal, 4.1% shooting) is likely out, leaving Tomas Hertl (0.88 xG/60, 5G, 17A, 1 HD goal, 4.3% shooting) as the secondary engine. Seattle counters with Shane Wright (0.77 xG/60, 8G, 10A, 4 HD goals, 10.4% shooting), who creates fewer chances but converts ruthlessly—his 10.4% shooting rate and 4 high-danger goals reflect lethal finishing. Berkly Catton (0.68 xG/60, 6G, 8A, 5 HD goals, 8.8% shooting) follows the same profile: low volume, elite efficiency.

Defensively, neither team hides weakness. Vegas' liability is Rasmus Andersson; Seattle's is Chandler Stephenson. Both teams sit at 0% high-danger-goal and 0% rebound-goal rates, removing screen-and-tip variance from the equation.

The model's logic: Vegas should win because it plays the better game. Whether PDO mean-reverts decides the outcome.