Dallas stole this one. New Jersey controlled 57% of expected goals and still lost 3-0—a variance so severe it amounts to competitive negligence.

The Devils held 52% of shot attempts and manufactured four high-danger chances to Dallas's two. But xG tells the grimmer story: they generated 3.05 expected goals and converted zero. Dallas generated 2.31 and buried three. In a game that should have favored New Jersey by the underlying metrics, the Stars executed ruthlessly.

Mikko Rantanen's wrist shot from the doorstep early in the third period shifted momentum. Dallas built on that high-danger chance; New Jersey couldn't answer.

J. Oettinger stopped everything he faced, finishing 30-for-30 and contributing +3.05 above expectation—the margin that separated Dallas's actual talent from its xG. J. Markstrom fielded a lighter workload at 25-for-28 but surrendered 0.69 goals more than expected, the byproduct of facing cleaner looks from a more efficient opponent.

Dallas did what teams dream about: dominated where it mattered, limited mistakes, let the goalie carry the load. New Jersey played well and lost anyway.

The model projects a 2.4–2.5 goal game with New Jersey at 51% win probability. This is about as close to a pick'em as you'll find.

Dallas arrives as the hotter team, 8-1-1 in their last 10. New Jersey's advanced metrics suggest otherwise. The Devils drive territorial play at 51.0% Corsi while Dallas sits at 47.0%. Dallas counters with a marginal xG edge: 51.0% to 50.0%. Neither is decisive.

The real story hides in variance. Dallas runs a 1.024 PDO—unsustainably hot goaltending and bounces combined. New Jersey sits at 0.971, flagging either a harder shot diet or simple bad luck. Both trends will regress; the timing is unknowable.

Dallas enjoys a three-day rest advantage over New Jersey's single day off. J. Allen faces the back-to-back for Dallas and may start fatigued or yield to a backup.

Jason Robertson anchors the Stars at 0.96 expected goals per 60 minutes with 25 goals, 21 assists, and a 12.4% shooting rate inflated by two high-danger finishes. Michael Bunting wheels at 0.83 xG/60 with eight goals, 14 assists, and 7.6% shooting—a volume player, not a clinical finisher. Timo Meier registers 1.08 xG/60 for New Jersey, the highest here, with 13 goals, 15 assists, three high-danger finishes, and a concerning 6.7% rate suggesting regression. Jack Hughes contributes 0.88 xG/60 with 18 goals and 17 assists at 11.5% shooting; one high-danger finish flags efficient but constrained volume.

Both teams show zero high-danger goal rate and zero rebound goal rate—a red flag for finishers or goaltending systems that lock down after the first chance. Dallas's Tyler Myers is a weak defensive link. New Jersey's Johnathan Kovacevic carries the same liability.

The Devils' possession edge and the Stars' recent pace create genuine tension. One regression, one hot goaltender, one moment—any shifts this from even to tilted. That's the coin.