The Blue Jackets defeated the Sabres 5–1 despite losing the expected-goals battle 3.76–5.78. Buffalo controlled 61% of xG at 5v5—a margin that should have meant a comfortable victory.

It didn't, because J. Greaves was exceptional. The Blue Jackets' goaltender stopped 32 of 33 shots (0.970 SV%), overperforming his expected save percentage by 4.78 goals. With a performance like that, Columbus didn't need to win the metrics; they just needed to capitalize when it counted.

Cole Sillinger's snap shot from the high slot (P3, 11:46) provided the opening salvo. Buffalo had the territorial advantage—11 high-danger chances to Columbus's 5—and yet converted none. The Blue Jackets, by contrast, found the back of the net in unconventional places: two from low-danger spots, four off the rush on 28 total rush chances. U. Luukkonen was flawless on high-danger opportunities (100% SV on 5 shots) but had no answer for Columbus's opportunism elsewhere.

The story of the game is written in the variance. CBJ overperformed their xG by 1.24 goals. Buffalo underperformed by 4.78. Combined, that's a six-goal swing toward Columbus. The Blue Jackets held 52% of shot attempts (35–33 on goal), which provided enough parity to stay in the game—but Greaves' brilliance and timely finishing on secondary chances are what won it.

This wasn't a clinic in advanced hockey. It was a master class in capitalizing on an opponent's waste.

Buffalo and Columbus are virtually indistinguishable on paper. The model projects a 2.5-2.7 goal spread with the Blue Jackets at 54% win probability—this is as close to a pick'em as exists.

Form masks deeper competitive balance. The Sabres have ripped off a 10-0-0 run, but that says nothing about talent relative to Columbus. At 5v5, the Blue Jackets edge Buffalo on both efficiency and volume: 52.0% expected goals and 51.0% Corsi against Buffalo's 50.0% and 49.0%. A coin flip, then, with underlying metrics backing it.

Buffalo's vulnerability is sustainability. The Sabres' 1.020 PDO is unsustainably hot; the Blue Jackets' 1.003 sits closer to normal, meaning Columbus will remain competitive once goal-scoring regression arrives.

Buffalo builds on Jason Zucker, who generates 1.07 xG/60 with 14 goals and 14 assists—six of them from rush chances at 15.1% shooting. Josh Doan follows with 1.01 xG/60, 15 goals, and 18 assists, but his 11.5% shooting rate masks unsustainable efficiency. Boone Jenner leads Columbus at 0.98 xG/60 with 11 goals and 21 assists. Mathieu Olivier provides secondary scoring at 0.89 xG/60 with 12 goals and 9 assists.

The defensive asymmetry matters only at margins. Jack Quinn is Buffalo's liability; Conor Garland fills that role in Columbus. Neither team converts high-danger chances reliably—both register 0% on high-danger and rebound scoring, a sign that neutral-zone execution, not defensive breakdown, determines play.

Take the Blue Jackets at 54% if analytics sway you. Take Buffalo if recent form does. This is a true toss-up.