The Islanders won 1-0 despite ceding 60% of expected goals to Edmonton (6.04–4.08). The Oilers also held 66% of shot attempts at 35–18. They lost anyway. This was a heist.

I. Sorokin (NYI) stopped all 35 shots, posting a 1.000 SV% and +6.04 against expected. He was perfect on high-danger chances. The Oilers couldn't convert theirs: 0 for 9 on high-danger attempts. The Islanders went 0 for 7. When both teams miss their looks, goaltending decides.

Edmonton generated quality elsewhere. The Oilers created 7 rebound opportunities and converted zero. New York created 3 and converted none. At 13:42 of the third period, Anthony Duclair's snap shot from the high slot ended the game. It was the only goal the Islanders needed.

C. Ingram (EDM) was solid at 17 for 18 (0.944 SV%, +3.08 above expected), but the volume worked against him. Edmonton owned nearly every metric—shot attempts, expected goals, high-danger chances—and had nothing to show for it. The Islanders stole this one.

The model favors Edmonton at 55% win probability in what it projects as a 2.5–2.8 goal game. That edge is rooted in Alberta's control metrics: the Oilers are winning the 5v5 xG battle 52.0% to 48.0%, a meaningful gap that suggests they're generating the better chances at equal strength. Edmonton's Corsi% (50.0% vs. 49.0%) reinforces that pattern, though the difference is tighter—a reminder that possession metrics and shot quality don't always align perfectly.

The more compelling argument for Edmonton surfaces in PDO regression. The Oilers are running a 0.982 PDO, which marks them as unusually cold from an efficiency standpoint. A team controlling play this well shouldn't see pucks bounce this badly. The Islanders' 0.993 PDO is closer to neutral, meaning neither team has luck advantage left to lose.

Goaltending introduces uncertainty. T. Jarry is on a back-to-back, creating the possibility of a backup appearance or a starter managing fatigue. I. Sorokin faces the same schedule pressure. If either lineup shifts, the injury calculus changes.

New York's offensive burden falls on Anders Lee (1.20 xG/60, 9G 20A) and Bo Horvat (0.87 xG/60, 18G 10A). Lee's 2 high-danger goals and 6.4% shooting suggest his goal rate remains unsustainably depressed relative to his chances, a canary for regression. Horvat has scored at a 12.8% clip on his looks—closer to sustainable—backed by 3 HD goals.

Edmonton's advantage rests on Zach Hyman (1.23 xG/60, 14G 13A) and Connor McDavid (0.92 xG/60, 25G 33A). Hyman's 1 HD goal on heavy volume signals he's out-scoring his underlying play, a note for regression risk. McDavid's 5 HD goals and 13.7% shooting represent elite volume and elite finishing, the kind of combination that doesn't reverse easily.

Defensively, each team has a liability. Alexander Romanov is the weak link on New York's blue line; Jake Walman is Edmonton's vulnerability. Neither team generates significant secondary scoring through high-danger play (both at 0% on rebound and HD-goal rates), which concentrates goal-scoring responsibility on the stars.

The regression story leans Edmonton's way: Hyman's efficiency normalizes downward, McDavid simply operates at his ceiling, and the team's cold PDO corrects upward.