The Defense Fixed Itself. That's the Problem.

The most dangerous number in Edmonton's offseason isn't the 82 points McDavid put up, or the cap commitment attached to Connor Murphy's blue-line contract. It's 1.7. That's how many fewer goals per game the Oilers surrendered in the final ten games of the season compared to the first twenty—a collapse followed by a correction, all within the same year, under the same coaching staff that no longer exists.

Here's why that matters, and why it cuts both ways. The late-season tightening to 2.3 goals-against per game is real evidence that a defensive identity is achievable with this roster. The organization clearly read it the same way, because the entire offseason logic—Babcock's hiring, Murphy's signing, the Nurse trade—flows from the conviction that the surge was a signal, not a fluke. I don't think that's wrong. The swing from 4.0 to 2.3 is the most statistically robust finding of Edmonton's season, and dismissing it as noise would be lazy. Something changed in those final games. The question is whether it was tactical, personnel-driven, or simply a schedule artifact that made everyone feel better about a lost year.

What the defensive improvement cannot answer, though, is the problem I've been writing about since July: the 13-22 record in one-goal games. Those two facts coexist uncomfortably. A team can tighten defensively and still lose close games at a 37% rate if the supporting cast can't execute in the moments that decide them. Bouchard's 46 points anchored the back end, but the production cliff behind him was steep, and Darnell Nurse—gone now to San Jose—was a minus-9 in 48 games while eating over 20 minutes a night. The defense got better late. It also wasn't good enough when games were on the line all season long.

Babcock inherits a roster that has demonstrated it can defend. What it has not demonstrated is that it can close. Those are related skills but they are not the same skill, and the offseason moves address the former more directly than the latter. Murphy adds structure on the right side. Dickinson's re-signing stabilizes center depth. Frederik Andersen brings the kind of playoff-tested goaltending that can steal a one-goal game—and that matters, because a veteran goalie who has been through deep playoff runs understands that the margin between winning and losing tight games is often a single save decision, not a defensive breakdown. That's the quiet bet buried in the Andersen signing.

My read is that the late-season defensive surge is a legitimate foundation, but it's being asked to carry more organizational weight than it should. A ten-game sample at the end of a season that was already over is encouraging. It is not proof. The real test for Babcock isn't whether Edmonton can hold a lead in November—it's whether the structure holds when McDavid and Draisaitl are being shut down in a tight game in April and the team needs someone else to find a way. That someone-else problem wasn't solved by the offseason. It was managed. The September 29 opener against Vancouver will tell us almost nothing. The one-goal games in February will tell us everything.