Columbus generated 55% of expected goals and controlled 55% of shot attempts—29 to Pittsburgh's 24—yet lost 4-3 in overtime. This was a goaltending and transition-efficiency victory, not a play-quality win for Pittsburgh.

T. Jarry stopped 1.02 goals above his expected save total; J. Greaves surrendered 0.72 below his mark. The two-goal swing between goaltenders in a one-goal game decided it. Columbus's attack was legitimate: eight high-danger chances versus Pittsburgh's seven, five rebounds generated (one converted), offensive pressure that should have produced more than three goals. Yet the Penguins converted transition chances at nearly double the rate—three goals on 22 rush attempts versus Columbus's two on 26—and K. Letang finished the game at 00:59 of overtime with a snap shot from the slot.

Z. Werenski led Columbus with a goal and assist in 26:06. S. Crosby (2G, 19:10) and K. Letang (1G, 1A, 22:33) drove Pittsburgh's offense. Neither Werenski's production nor Columbus's statistical dominance changed the result. Pittsburgh's formula was clear: let the opposition control the game, execute ruthlessly in open ice, and trust the goaltender. This is NHL variance—sustained pressure and quality chance generation fall short when the opponent finishes transitions.

The model projects a 2.5-2.7 goal spread favoring Pittsburgh, with Columbus at 53% win probability—close enough to a pick'em that the edge is negligible.

Neither team has momentum. The Blue Jackets sit at -7 goal differential over their last 10 games; the Penguins are at -5. Yet at five-on-five, the underlying possession numbers are nearly matched. Columbus edges Pittsburgh in expected goals (52.0% to 51.0%) and Corsi percentage (51.0% to 50.0%), but both margins are marginal. Pittsburgh's PDO sits at 1.012—a sign that shooting luck has been generous—while Columbus runs 1.003, suggesting closer alignment with underlying performance.

The goaltending situation could prove decisive. J. Greaves and T. Jarry are both on back-to-backs, raising the risk of either team deploying a backup or leaning on a fatigued starter in the second half.

Pittsburgh's Attack and Absence

Rickard Rakell carries the forward load: 13 goals on 12.1% shooting, with 1.01 expected goals per 60 minutes and 3 high-danger finishes. That shooting percentage is unsustainable—expect regression—but it proves the offense exists beyond noise. Filip Hallander, who grades at 1.28 xG/60, has been recalled from a conditioning loan with the AHL Penguins for further evaluation and testing. His status remains uncertain, which matters given the offensive depth he represents.

The Penguins' structural problem is brutal. They register 0% high-danger goal rate and 0% rebound goal rate—a sign that goals cluster in low-probability moments. Caleb Jones anchors their defense's weakest point.

Columbus Counters

Boone Jenner is Columbus's primary threat, posting 11 goals and 0.98 expected goals per 60 on 9.6% shooting. Mathieu Olivier provides depth: 12 goals at 13.3% shooting, with 1 high-danger finish and 0.89 xG/60. Neither reaches Pittsburgh's individual volume, but Columbus distributes its scoring more evenly.

The Blue Jackets face the same architectural flaw—0% high-danger and 0% rebound goal rates—suggesting both squads rely on opportunism rather than system. Conor Garland marks Columbus's defensive liability.

The Verdict

Columbus's marginal advantage in underlying metrics cannot overcome the teams' identical reliance on low-probability finishes. Pittsburgh's shooting luck will normalize. That 53% edge for Columbus reflects parity, not proven superiority. Expect a game matching its projection: tight, decided by which team converts the few real chances, settled by goaltending discipline on a night when both nets should be tired.