Carolina's dominance on the underlying numbers—66% of expected goals—made the 6-3 scoreline almost conservative. The Hurricanes held 61% of shot attempts and built a 5.62–2.89 advantage in expected goals, the kind of disparity that usually yields larger margins.

Buffalo's goaltender A. Lyon kept the margin respectable. He saved 1.6 goals above expected, a herculean effort that prevented the deficit from widening further. Meanwhile, P. Kochetkov faced an easier task: he was perfect on high-danger chances, turning away all five.

The shot geography revealed Carolina's advantage. The Hurricanes generated 0.078 expected goals per shot—premium scoring areas on most attempts. Buffalo managed 0.058, a better positioning than their volume alone suggested, yet insufficient against Carolina's efficiency. The Hurricanes' rush game was particularly lethal: they converted 5 goals on 28 rush attempts, while Buffalo managed only 3 on 22.

High-danger chances finished 8–5 in Carolina's favor, reflecting their dominance in the most critical scoring areas. Buffalo couldn't convert on 5 high-danger opportunities—an anomaly that won't repeat.

The underlying numbers aligned perfectly with the final score. When one team controls shot volume, shot quality, and proximity to goal by this margin, defeat becomes inevitable. A. Lyon's exceptional performance was merely a plot twist, not a reversal: individual excellence cannot overcome systemic domination.

The model projects a 2.4 to 2.9 goal game favoring Carolina at 58% win probability—a lean that rests on territory, not luck.

The underlying advantage

Carolina owns the 5v5 shot table decisively. The Hurricanes generate 56.0% expected goals and control the Corsi at 60.0%, meaning they should dictate play tempo and location. Buffalo sits at 50.0% xG and 49.0% Corsi, a gap wide enough to predict sustained pressure.

What does this mean? Carolina will spend most of the game in Buffalo's zone. Buffalo wins some games with this script through goaltending or timely finishing, but the model expects it to lose this one.

The PDO wrinkle

This is where regression enters. Buffalo's PDO of 1.020 means they're converting at an unsustainable clip—they've extracted more value from their shots than underlying metrics predict. Carolina's PDO is 0.984, a cold indicator suggesting the Hurricanes are underperforming their quality. The model bakes in a normalization: expect Buffalo's shooting luck to cool, expect Carolina's to climb.

Goaltending pressure

Both F. Andersen and U. Luukkonen face back-to-backs. Andersen could see the bench in favor of a fresher backup; the same risk applies for Luukkonen. If either plays tired, a deeper goal deficit becomes plausible despite the quality gap.

Who decides it

For Buffalo, Jason Zucker drives offense at 1.07 xG per 60 minutes with 14 goals and 14 assists, though his 15.1% shooting rate is elevated—expect some regression there. Josh Doan follows at 1.01 xG/60 with 15 goals and 18 assists. Jack Quinn represents a structural defensive weakness.

Carolina counters with Jackson Blake at 1.06 xG/60, 14 goals and 19 assists, and Seth Jarvis at 0.98 xG/60 with 17 goals and 12 assists. Charles Alexis Legault is Carolina's defensive liability.

Bottom line

Carolina should control this game. The model respects that dominance more than it respects Buffalo's recent output. A Hurricanes win happens because of possession advantage, not because of any special alchemy. The 58% probability is cautious, not bullish—it accounts for the chaos of one game and the possibility of a hot goalie bailing out a struggling team.