New Jersey beat Columbus 3–2 despite losing the xG battle 5.01–3.21. Columbus controlled 61% of the expected goals—this was a theft.

The Blue Jackets dominated the metrics across the board. They held 52% of shot attempts (35–33), created six high-danger chances to New Jersey's four, and generated the scoring opportunities that should have won them this game. Yet they left empty-handed, undone by both misfortune and precision from the opposite end.

Columbus underperformed their xG by 3.01 goals. That is the margin of defeat in a game where they were the better team. J. Greaves made 30 of 33 saves (0.909 SV%), a respectable outing that masked his team's execution in front of the net.

J. Allen was the difference. The New Jersey netminder stopped 33 of 35 shots (0.943 SV%), posting a +3.01 performance against expected goals. Against the run of play, he was exceptional.

The turning point crystallized early in the third. L. Hughes unleashed a snap shot from the high slot at 4:45—the kind of chance that rarely beats a hot goalie, but this time it did. That goal shifted momentum in a game Columbus should have controlled.

New Jersey capitalized on transition. The Devils scored three goals on 27 rush chances, converting movement into results. Columbus generated nine rebound opportunities versus five for New Jersey, but converting twice was not enough against such superior shot creation.

This was not a matter of luck. Columbus simply did not bury their chances, and Allen elevated to steal a win his team had no right to take.

The model projects a 2.4–2.7 goal range with the Blue Jackets at 54% win probability—about as close to a pick'em as you'll find. Neither team has earned separation.

Both are underwater. Columbus sits -5 in goal differential over its last 10 games; New Jersey at -6.

At 5v5, the underlying numbers are locked. The Devils own 50.0% expected goals and 51.0% Corsi; the Blue Jackets 52.0% xG% and matching 51.0% Corsi. New Jersey's PDO of 0.971 shows it's running cold—a team due for positive regression. Columbus's 1.003 PDO is the inverse: a team that's gotten lucky. One will revert.

The Devils hold a structural advantage: three days of rest versus Columbus's one. J. Greaves may see a backup in goal, or a tired starter.

Timo Meier carries New Jersey's offense at 1.08 expected goals per 60 minutes, with 13 goals and 15 assists. His 6.7% shooting rate is suppressed—room for regression upward. Jack Hughes, at 0.88 xG/60, owns 18 goals and 17 assists on an 11.5% clip, closer to sustainable. For Columbus, Boone Jenner leads at 0.98 xG/60 with 11 goals and 21 assists, but his 9.6% shooting suggests he's been unlucky. Mathieu Olivier adds 0.89 xG/60, 12 goals, and 9 assists on 13.3% shooting—the healthier rate.

Neither team is generating secondary scoring. Both sit at 0% high-danger and rebound goal rates. New Jersey's defensive liability is Johnathan Kovacevic; Columbus's is Conor Garland. The team that breaks through first on its star power likely wins.