The Senators owned this game by every measure except the scoreboard. New Jersey won 4-3 despite trailing in expected goals 4.18–2.68, a 61–39 split that should have belonged to Ottawa. The Senators controlled 54% of shot attempts and 7 high-danger chances to 2. The Devils took the victory anyway.

This was goaltending variance crystallized. J. Markstrom's 0.921 save percentage (35 of 38) ran 1.18 goals ahead of expectation. L. Ullmark's 0.875 mark (28 of 32) ran 1.32 behind. The two-goal swing between the pipes was Ottawa's game surrendered.

The fracture came early in the third period. Cody Glass fired a wrist shot from the slot at 12:24 and the game's momentum shifted. Before that moment, Ottawa had dictated play. After, the Senators chased.

Even Ottawa's finish couldn't compensate. They scored twice from 26 rush attempts while New Jersey converted four from the same 26 chances—a perfect inversion. The Senators created five rebound opportunities, converting one. New Jersey created three and scored zero. The Devils still won.

Ottawa's power play was elite: three goals on 15 shots. Dominance. It mattered less than it should have. The Devils took what they didn't deserve and made it stick. The Senators got robbed.

The model projects a 2.3–2.7 game, favoring Ottawa at 57% win probability—despite New Jersey's PDO regression potential.

Ottawa owns the underlying play. At 56.0% 5v5 xG%, the Senators generate better chances than New Jersey's 50.0%. That advantage extends to possession: 53.0% Corsi% for Ottawa versus New Jersey's 51.0%.

Both teams are currently struggling. New Jersey has surrendered a -10 goal differential over its last 10 games; Ottawa is -5. The Devils' 0.971 PDO signals they're running cold, primed for positive regression. Yet Ottawa's structural advantage is large enough that even accounting for New Jersey's likely improvement, the Senators remain the lean at 57%.

The offensive edge is visible. Timo Meier operates at 1.08 xG/60 with 13 goals and 15 assists, backed by 3 high-danger goals at 6.7% shooting. Jack Hughes carries 0.88 xG/60 with 18 goals and 17 assists; his 11.5% shooting efficiency masks his moderate volume, and he has 1 high-danger goal. Brady Tkachuk counters at 1.19 xG/60 with 11 goals, 20 assists, 2 high-danger goals at 7.5% shooting. Dylan Cozens follows at 1.00 xG/60 with 13 goals, 15 assists, 1 high-danger goal at 9.5% shooting.

Defensively, both teams sit at 0% high-danger goal rate and 0% rebound goal rate. Johnathan Kovacevic is New Jersey's weak link; Nick Jensen is Ottawa's. In a tight game, those spots matter.