Columbus dominated Utah on the underlying metrics — 66% of 5v5 expected goals (4.55–2.37) and a 7–2 high-danger-chance edge — yet the 3–2 overtime result masks the thoroughness of the Blue Jackets' control. V. Vanecek's exceptional goaltending kept the Mammoth in a game the metrics suggested Columbus should have won decisively.

Vanecek saved 1.55 goals above expected, essentially stealing one. That differential — one goal beyond an average performance — explains why Columbus's dominance didn't translate to a larger margin. The Blue Jackets generated higher-quality chances throughout, averaging 0.062 expected goals per shot versus Utah's 0.036, and attacked the crease relentlessly. They crashed for 6 rebound opportunities, converting one; they were lethal on the rush, scoring 2 of 33 chances; their power play converted 2 on 13 shots for 1.87 expected goals.

Yet Columbus still underperformed their expected goals by 1.55. They had the looks; finishing faltered. J. Greaves steadied the Blue Jackets in net with a 0.926 save percentage on 25 of 27 stops — solid, if not spectacular.

Dmitri Voronkov's overtime tip-in at 01:01 broke the deadlock, arriving from the high-danger areas Columbus had dominated all evening.

The model projects a 2.5-2.7 goal range and gives Utah 53% win probability. This is as close to a pick'em as you'll find.

The teams are genuinely equivalent. Columbus and Utah both generate 52.0% expected goals at 5v5; Utah edges possession at 53.0% Corsis versus Columbus's 51.0%. Both sit at exactly 1.003 PDO, indicating neither is running hot or cold from a luck standpoint. The predictive numbers say this should be a coin flip—and the model agrees.

Goaltending adds uncertainty. K. Vejmelka starts a back-to-back for Utah, and E. Merzlikins does the same for Columbus. One or both could see a backup, or face a fatigue-taxed starter. That's an X factor neither team wants in a close game.

For Columbus, Boone Jenner is the fulcrum. He's generating 0.98 xG per 60 with 11 goals and 21 assists; his 2 high-danger goals on a 9.6% shooting rate suggest he's due some regression or breakout—either direction moves the needle. Mathieu Olivier at 0.89 xG per 60 carries 12 goals and a hot 13.3% shooting clip, which Utah's defense needs to respect.

Utah counters with Dylan Guenther, who leads decisively: 1.02 xG per 60, 28 goals, 15 assists, and 3 high-danger tallies at a 16.8% shooting rate. He's not running lucky; he's substantially better at finishing chances. Daniil But (1.01 xG per 60, 2 goals, 2 assists, 1 high-danger goal) is the secondary concern—promising metrics on a light goal total means more finishing is coming.

The tactical detail: both teams sit at 0% high-danger goal rate and 0% rebound goal rate, suggesting defensive systems that eliminate second chances. That tightness works against volume scoring but rewards opportunism. Conor Garland is Columbus's weak link on the back line; Maveric Lamoureux is Utah's. Both are attack targets.

This isn't a guess. The models say coin flip, the matchups say coin flip, and the goaltending situation adds chaos. Utah's edge comes from better production at the point of attack, but Columbus's chance quality is nearly identical. Defer to the 53%.