The Discourse: Thomas Harley for Norris, Stankoven scratched, and another successful weekend

Hockey's back!

If you’re something of a hockey snob like me, you couldn’t help but feel almost let down by the NHL action this weekend as compared to what the 4 Nations offered. But I’m also glad the 4 Nations is over. And I’m glad the Dallas Stars didn’t get that memo about “lesser hockey.”

The weekend wasn’t without a few scares, but Dallas left no doubt about who the better team was.

Starting with New Jersey, it was easily the cleaner win of the back-to-back. The Stars never trailed in a game that probably would have had more scoring if it weren’t for the respective goaltenders. Matt Duchene and Wyatt Johnston both had multipoint nights, cementing themselves as Dallas’ two best forwards (or so it seemed until Sunday). It felt fitting too, because the more I watch Johnston, the more he appears as if he’s trying to replicate Duchene’s movement. Johnston’s skating wasn’t an asset as a prospect, but it’s turned into one as he’s improved in this regard.

The Islanders game was a bit wonky, for reasons we’ll dig deeper into (and no thanks to the hit on Lian Bichsel), but for the most part, the Stars played well given the circumstances. The neutral zone remains an issue, but as long as Miro Heiskanen is out, it’s not like it’s gonna get better, so I won’t harp on this point.

One player that deserves a shoutout is Matt Dumba. He had an actively good game versus New Jersey — the kind of game that felt pulled from an entirely different player — and versus New York, it was great to see him come to Bichsel’s aid after the cheapshot by Casey Cizikas. One good weekend doesn’t erase his body of work thus far, but it’s the kind of play you hope he builds on. While he’s bound to be the odd man out whenever Heiskanen returns, it’s nice to see him play within his role, however fleeting.

By no player was more essential to Dallas’ win on Sunday than Jason Robertson, who is aiming to catch Duchene as Dallas’ point leader. We’ve come a long way from “he’s probably still recovering that offseason surgery.” Robertson is as dialed in as he’s ever been, which is important, because Dallas doesn’t have a scoring problem so much as they have a “who can take over the game?” problem. It’s the age-old debate going into the playoffs about what kind of team tends to win. With games like this, we’re getting closer to knowing who has the power of one.

While it’s not inconceivable for Dallas to catch Winnipeg, it’s certainly less likely; which further sets the stage for a first round matchup with Minnesota. (Colorado is only 5-5 in their last 10, and lost both games this weekend.)

Thomas Harley in the Norris conversation?

Yes and no. According to Dom’s Net Rating, which works a little like Wins Above Replacement, Harley belongs in the Norris conversation. That begs the question: is Harley really as good as Josh Morrissey? (Keep in mind, as these ratings are made with regard to awards, they are highlighted as totals rather than rates. Hence the rating of Quinn Hughes, whether you agree or not, who probably leads the pack if he hadn’t missed so much time.)

Data and visual courtesy of Dom Luszczyszyn (NHL Awards Watch).

It’s always important to stress what these numbers mean. For example, Dom’s Net Rating was born out of basketball’s Game Score concept: a kind of catchall ‘who had the best night?’ stat. A way of thinking about this stat is by considering what a player’s plus-minus would look like if they included the action from shift to shift rather than just goal to goal. In other words: a cookie for winning a faceoff or not, for being on the ice for a shot attempt or not, a cookie for drawing a penalty versus taking one, and so forth, as I’ve tried to explain to layman’s terms.

The point of this preamble is that casual fans always love when a single value stat makes a player they root for look elite, but they’re hardboiled detectives when the stats make a player they root for look like trash. So what are we to make of this? Is Harley really on the level of a Josh Morrissey?

The first thing I’ll point out is the irony: Morrissey didn’t hit those numbers until last season. His elite level of play is a recent development. We’re not talking about Victor Hedman, or Cale Makar, who have had sustained windows of elite play. Players change all the time. And I think that’s what we’re seeing with Harley. He has solved that riddle of steel I spoke so pretentiously about. For reference, because it’s the only prescient thing I’ve ever written:

A lot will be said about Harley playing in Miro Heiskanen’s shadow. You might even wonder what Harley would look like without Heiskanen. Dom’s model, like all the best models, accounts for quality of teammates and competition as best they can to give a single-value stat that tries to isolate player value. However, I think these arguments expose paradoxes rather than imprecision. Maybe Harley’s bottom line value is boosted by Heiskanen, because what player wouldn’t be, in the same way that Makar’s relationship with Nathan MacKinnon is quite symbiotic? But at what point is Harley’s bottom line value also a reflection of the chemistry he generates with others? At what point does Harley’s bottom line value mirror what he’s capable of when everything coalesces?

This is what it means to engage with the data, rather than blindly advocate or dismiss them. How can a player with 174 games suddenly be in the conversation for the Norris? Well, maybe he’s that good. Maybe his peers are falling behind. Maybe it’s a perfect storm kind of year. Whatever the mitigating factors are, they don’t minimize the very real impact he’s having when it comes to shot impacts, and shot quality performance. Part of the value in numbers is the value in a new language: in this case, the language of Harley compared to his peers. At the 4 Nations Face-Off, he was very clearly one of Canada’s top defenders. Maybe that says something about his place on the spectrum of the game’s best defenders?

This is not to say I accept this wholesale. Even Dom has a massive neon sign atop his article that everyone in the comments section often ignores: “charts don’t tell the whole story.” In my experience, it’s the people who seek to ignore them completely that don’t bother reading the book from start to finish. But it does raise questions. After all, this data is a snapshot of who Harley is, from shift to shift, for this season. Is this a sign of things to come? Has Harley arrived in a way that puts him in the top 10? Is this just a lucky year? I think these are reasonable questions.

But no matter the answer, as I’ve always said…Thomas Harley: elite.

Stankoven scratched

Pete DeBoer was very candid about the reason: we have a deep forward group, so if you’re not standing out, somebody else gets to stand in.

Broadly speaking, I don’t have a problem with this. Watching Stankoven—something I’ve been pretty explicit about in the past—it’s clear that he’s not comfortable all the time. While he’s getting shots off, he’s looked skittish to these spreadsheet-watching eyes. He’s too quick to release, sometimes at sharp angles or low danger areas, and I swear by his abrupt lack of physicality. Last year, Stankoven was delivering hits and cutting rug. This year, he’s been knocked off the puck, readily pokechecked, and sometimes even gets bullied. Do I think these are long term problems? Not at all. But there’s a layer missing in Stankoven beyond simply being snakebit, and DeBoer is clearly catching that.

However, does that make him the odd man out? Following the Harley-For-Norris logic, no. In point of fact, out of all the players who have bounced around the bottom of the lineup, Stankoven should be the last one to draw out.

Does one chart tell the whole story?

No. But he still leads all of these players in points by a decent margin, in addition to having a superior shift-to-shift impact. Two things can be true at once: Stankoven is struggling, and Stankoven is not struggling. In this case, he’s struggling to score and make the impact he’s shown he’s capable of, but he’s not struggling to help Dallas control territory. This is evident in his tracking data per Corey Sznajder.

That’s a hell of a profile.

This is not to scream #FreeLogan. He’s truly fighting some puck demons, and DeBoer is leveraging the oldest trick in the book when it comes to any team’s youngest players. I don’t believe “regression” is gonna be the silver bullet in this case because a lot of factors are at play. But I do believe Stankoven is due in some capacity. I mentioned on Twitter that I’d like to see him take some shifts at center, as that was his position with the Kamloops Blazers. As a center, he was known for his faceoff prowess and leveraging his low center of gravity. We’re getting ahead of ourselves, obviously. Stankoven will be fine, but his road to becoming a key part of Dallas’ run got a bit tougher.