The New Jersey Devils have been off to an incredible start: their 8-1-0 record places them atop the entire National Hockey League (NHL). While basically everything — aside from injury luck — has gone their way, there’s arguably been no individual player more important than goaltender Jake Allen.
Related: Devils’ Assistant Brad Shaw Making Immediate Positive Impact
Allen’s recent win over the Colorado Avalanche moved him to 5-0-0 with a 1.93 goals against average (GAA) and .923 save percentage (SV%). This is even more remarkable when you consider that Jacob Markstrom missed six games, prompting Allen to become the de facto number one. He rose to the challenge…and then some.
Allen’s Numbers Can’t Be Ignored
While a backup netminder typically starts ~30 games, it’s time to stop ignoring the eye-popping numbers: Allen has legitimately been one of the best goaltenders in the NHL since last season and deserves more playing time. Now, before we get into those numbers, it’s important to note that this has nothing to do with Markstrom’s play. He’s a very good netminder. This is simply a luxury the Devils gain by having two great options in the crease.
Last season, in a 31-game sample size, Allen’s +18.4 goals saved above expected was the ninth-best mark out of 103 NHL goalies. (via MoneyPuck) However, his counting stats never really warranted conversation for more than a backup opportunity: his save percentage of .908% was great, but nothing earth shattering.
How could his advanced numbers be so elite…but his save percentage unreflective of that? Well, the Devils’ defense strangely played way worse in front of Allen; there doesn’t seem to be any reasoning behind it. It seems more coincidental than anything else. This season, that issue has been righted and the counting stats reflect that.
To get a better glance at how impressive his advanced numbers were in 2024-25, per 60 minutes, he stopped +0.635 goals saved above expected (GSAx). That was good for second in the league among all goaltenders to play 15+ games, tied with Connor Hellebuyck and only trailing Anthony Stolarz.
To tie it all together and add some perspective: this season, Allen has stopped +0.597 GSAx per 60, which while impressive, is ~6% fewer than last season…yet he’s undefeated with a .923 SV%. That speaks to the level he can play to when the defense is doing their part. (Side note: the defense has looked promising early on under new assistant Brad Shaw, who players and coaches alike have raved about.)
Sure, with more starts, Allen’s numbers could regress a little. But a now ~40 game sample of being elite is nothing to scoff at. In fact, not only were Allen’s advanced numbers and counting stats better than Markstrom’s last season, but they have the exact same career save percentage (.908%).
They’re both 35 and have had injury troubles in recent seasons; having the luxury of evening out the load will induce healthy competition while not overworking either one come playoff time.
Markstrom struggled a bit to start the season with a 3.89 GAA and .845 SV% in three games. But it shouldn’t be a cause for concern — he looked great during preseason, and his track record proves that he’s capable of quickly bouncing back from rough stretches.
If they’re going to eventually get to a more even split, it will take some time as Allen has played a ton in Markstrom’s absence.
After their most recent home game, head coach Sheldon Keefe said, “Allen’s been unbelievable (…) At the same time, we also have to manage his workload. We just touched on the [demands of] the schedule, and now we’re going to travel with Colorado. So you know, it’s good to have goaltending depth, and it’s good to have Marky coming back at this time. With the amount of schedule here, we have to manage that, or we’ll have injured goalies all season long. We don’t want that.”
Last season, Markstrom started 49 games while Allen started 29. What a more evenly matched scenario looks like will depend on factors like health and recent performance, but 29 is too few for Allen. He’s way too good.
