Pittsburgh controlled the play so thoroughly that a 5-4 overtime result feels generous to Columbus — the Penguins owned 64% of 5v5 expected goals (4.63–2.64), a gap that should have buried the Blue Jackets.

What kept the game close was simple variance. The Blue Jackets overperformed their xG by 1.36 goals, scoring three times from low-danger spots that goaltenders are expected to stop. Against that luck, A. Silovs underperformed, surrendering 1.36 more goals than his workload deserved; his 0.840 SV% on 25 shots masks how thoroughly Pittsburgh controlled the ice. J. Greaves, meanwhile, delivered (0.884 SV%), but his team never generated enough possessions to cash in.

The possession gap told the full story: Columbus managed 37% of shot attempts (25–43), a stranglehold that explained the offensive drought. Pittsburgh was ruthless on transition, converting three rushes from 39 attempts — the kind of efficiency that closes games. Their power play was dominant too, going 2-for-13 (0.52 xG).

Sidney Crosby ended it with a backhander from the doorstep at 02:22 of overtime.

The model projects a 2.5-2.7 goal game, giving the Blue Jackets 53% win probability. This is about as close to a pick'em as advanced metrics get.

The underlying talent is evenly distributed. Pittsburgh owns 51.0% expected goals share at 5v5 while Columbus posts 52.0%. Corsi splits are nearly identical at 50.0% for the Penguins and 51.0% for the Blue Jackets. Neither team's PDO (1.012 for Pittsburgh, 1.003 for Columbus) shows meaningful luck variance—both are skating within a whisker of sustainable performance.

The goaltending situation tilts toward chaos. J. Greaves faces back-to-back duties for Pittsburgh, while S. Skinner does the same for Columbus. Expect either a backup start or a noticeably fatigued starter from one or both sides.

Pittsburgh's offensive burden falls on Rickard Rakell, who carries 1.01 expected goals per 60 minutes and has banked 13 goals and 13 assists with 3 high-danger chances. His 12.1% shooting rate suggests some unsustainability, meaning regression is likelier than continued outperformance. Filip Hallander (1.28 xG/60) was a secondary threat but has been recalled from the AHL for further evaluation and testing, effectively removing him from this matchup.

Columbus counters with Boone Jenner, whose 0.98 xG/60 places him in that high-volume, scoring-relevant band. He owns 11 goals and 21 assists with 2 high-danger chances, and his 9.6% shooting rate leaves less room for positive variance than Rakell enjoys. Mathieu Olivier (0.89 xG/60, 12 goals, 13 assists, 13.3% shooting) offers secondary thrust and a shooting percentage that screams unsustainable.

Both teams' goal-scoring machinery has flatlined on high-danger and rebound opportunities—each sits at 0% rates, a red flag suggesting neither has converted what should be their easiest looks. Pittsburgh's defense features Caleb Jones as its weakest link; Columbus will lean on Conor Garland as theirs.

This game lives in true coin-flip territory because nothing separates the teams in any meaningful dimension. The slight model edge to Columbus reflects nothing more than statistical noise dressed as prediction.