The Gap Between the Record and the Reality Is Already Familiar
Last season ended with New Jersey holding a 44–41 record and a negative goal differential. That combination—winning record, goals against exceeding goals for—is the statistical equivalent of a house built on sand. The Devils knew it. The front office knew it. And yet three games into 2026–27, the gap between what the record says and what the underlying numbers say has already reopened to 6.4 win points.
That number is the one worth sitting with. New Jersey is winning at a 52% clip while their goal differential suggests they should be winning at roughly 45%. The 7-point spread isn't a fluke of sample size—it's a structural signature, and it's the same signature that defined last season's entire arc. A team that genuinely outplays its opponents doesn't need to outperform its goal differential to win. The Devils are doing exactly that, again, in the first week of October.
The mechanism driving it is equally familiar: New Jersey is winning one-goal games at a 65% rate. That's elite execution in the tightest moments, and it is also the most volatile thing a team can depend on. Close games are decided by bounces, by a goaltender's best save of the night, by a power play that converts when it has to. Sustained 65% performance in one-goal situations is historically rare because the variables that produce it are not fully within a team's control. Last season, the Devils rode that rate all the way to April—and still finished with a negative differential. The margin was always thinner than the record suggested.
What makes this version of the problem more acute is what's changed around the edges. Simon Nemec, who posted a plus-6 in 37 games last season, is gone to Calgary. His departure didn't just cost the blue line a point producer—it cost them one of the few defenders who was consistently on the right side of the ledger in a season where most of the roster ran negative. Luke Hughes leads the defensive corps now, and his offensive contributions are real, but his minus-9 from last season tells you something about the structural exposure behind him. A team leaking goals and winning close games cannot afford to lose the defenders who were helping contain the leak.
The goaltending question compounds everything. Sunny Mehta told reporters on September 18 that he's comfortable with the situation in net—and I'll take him at his word that he's made his peace with it. But comfort is not the same as resolution. Jacob Markstrom was traded to Florida. The veteran alternatives who might have filled that void have largely evaporated. Three games is nowhere near enough data to know whether the Devils have found something workable between the pipes, and the one loss—a 4–3 defeat to the Rangers on the 24th—is a reminder that close-game execution has a ceiling when the goaltending is uncertain.
The offensive trend line adds a third layer of concern. Scoring has dropped to 2.6 goals per game over the last ten contests, down sharply from the prior stretch. A team that depends on one-goal wins to pad its record cannot afford to score less. The margin in a one-goal game is, by definition, one goal. When your offense is trending toward 2.6 per night, every defensive breakdown becomes a potential loss, and every win requires near-perfect situational execution. That's a hard way to build a sustainable season.
My read is this: the Devils are not a bad team. Hischier, Bratt, and a healthy Jack Hughes give them a legitimate top-six. Dawson Mercer's plus-8 last season suggests there are players on this roster capable of being on the right side of the ice. But the organizational pattern—outperforming the underlying numbers through close-game execution, posting a winning record over a negative differential—has now appeared in back-to-back seasons. At some point, that stops being a coincidence and starts being a business model. And business models built on variance eventually get audited.