How Does it Affect the Stars? Moritz Seider and the 'analytics' debate, Elias Lindholm to Vancouver, and Igor Shesterkin's struggles
But first! A Thomas Harley discussion.
I would have had this up for everyone yesterday but I was in Austin, first to take care of some family, and then to watch Elder open for Tool. Yes, Elder was the main event for me. I mean, listen to these mfers. Tool is great, and when I was young, they were one of my favorite bands. But it sucks going to see a band and having to pay two hundred times what you would normally pay just because of who they’re opening for. Also, they only got a 26 minute-set! Some bullshit is all.
But nobody came here to talk about music. We came here to talk about the Stars. “But it’s All-Star week.” Where the hell have you been? Sean Shapiro and Dimitri Filipovic put out an absolute banger on Tuesday about Thomas Harley, mixing high level analysis via stats and various tape with a franchise-focused scope. (Sean’s been a busy boy this week, hashing out a piece on Jason Robertson, and rocking out with his feature out on Esa Lindell at D Magazine.)
One of the things that struck me about the EPRinkside piece was the brief discussion about Harley’s personality. The other day we talked about why Thomas Harley, who may well end up becoming the best defensemen from that 2019 draft (more on that in a bit), fell given his obvious physical tools against the backdrop of defensemen picked higher who had less. I’m not comfortable speculating, but knowing a thing and predicting a thing are two very different things. Is it possible that Harley, given his specific personality, just didn’t have a good interview?
I’m not interested in that answer because we’ll never know. But we can speak to previous draftees. Daniel Sprong famously fell because of his interviews. Jesse Puljujarvi fell because of something even stranger (a gift to Columbus in retrospect) — if true of course. Redditors have a running list, and while we can’t speak to specifics, we do know that the NHL draft interview process can be extremely bizarre. People like to say that the draft is a crapshoot, but a lack of mathematical certitude isn’t the same as random. Who knows. Maybe injecting less chaos into the draft might one day help.
Regardless, Dallas has a Harley. And they’re a lot better for it.
What Moritz Seider tells us about analytics
There’s always one. Or two. Or maybe a lot. But this year the Analytics Dud vs. Eye Test Star is Detroit defender Moritz Seider. In a nutshell, how does such an obviously good defensemen have such obviously poor stats?
Prashanth Iyer has a theory: historically tough defensive zone usage.
I was looking through PuckIQ’s QoC stat, and I couldn’t find anyone who comes close to the almost-eight minutes Seider plays versus elite competition. That’s nearly half his EV minutes against killers, to combine with the defensive zone starts. For perspective, Miro Heiskanen plays well less than seven minutes versus elite competition per game, and that leads the team.
There’s a wrong way to have this discussion, which is the usual variation of “well that’s because stats don’t tell you everything.” Or some nerdonaut saying that “well maybe Seider is just bad.”
The right way is to discover the right language for the discussion. For example: is there another kind of information that can clarify Seider’s performance? Through what lens do we evaluate Seider when we watch him? Questions like these can lead to model-makers creating new ways to measure different things, like off-ice impact. (Micah Blake McCurdy did some preliminary work on this concept in 2021.) And they can call into question whether or not we see what we want to see. We may not get answers, but we’re not owed them. We can only update the language. Information is not synonymous with value just as knowledge is not synonymous with understanding. If analysis is to mean anything for us in the context of hockey, it’s to respect what we don’t know, whether it’s the description that a mathematical model seeks to define, or the performance that a player effectuates.
Interestingly, Harley doesn’t have a strong defensive impact looking at goals above replacement models. But his transition data is promising.
I asked Corey Sznajder about it on his Substack, and well — read on!
Elias Lindholm is a Canuck
For Andrei Kuzmenko, Joni Jurmo, a 2024 fourth rounder if Vancouver makes it to the Conference Finals, a 2024 first, and one, Hunter Brzustewicz. Personally, I think that’s nuts, especially given what we know about Lindholm. But also, I wrote about Brzustewicz ahead of the draft for a reason: I think he’s legit.
But one thing is for sure: Vancouver just got better. While I don’t believe Lindholm will ever be worth what he’s gonna make next season, we are talking about a strong two-way center, who — whether causation or correlation — has proven he can also produce with elite linemates. The Canucks have the juice to maximize his skillset, and it helps fortify their team defense, which is otherwise lacking on the blueline.
The obvious effect on the Stars is that Vancouver is a potential playoff opponent. While I still don’t believer in their outright legitimacy, there’s no way they can be dismissed. In addition, Dallas just doesn’t have much luck against them. And that’s not counting other potential adds (by giving up Kuzmenko, Vancouver actually added some cap.)
The other effect is price. If Tanev costs something similar, what is the Dallas equivalent of Brzustewicz? (Tom Willander is their best defensive prospect.) I don’t know. I suspect a team might find a Dallas 2024 first, Christian Kyrou, and Ty Dellandrea intriguing, but Dallas doesn’t have a cap dump to make it worth it. So if Jim Nill swings big, he’ll need another team to be involved.
Igor Not-Quite-Shesterkin
Shesterkin has been the worst goalie in the month of January with an .863 save percentage. Bet you didn’t have that on your bingo card did you? In some ways it should be reassuring for Stars fans who worry about Jake Oettinger. Or maybe not. Regardless, I’m not comfortable being extra critical of Oettinger, who has given Dallas far more quality performances than bad ones. But I understand the timing. Why did Oettinger have to have a down year with this roster?
However the numbers break down in the end, the hope is obviously that Oettinger regains his form when the chips are down. If so, it’s gonna be hard to stop this Dallas team, regardless of any trade deadline moves.
To get it all started, Dallas is set to face the Buffalo Sabres. The Sabres have been no joke lately. Ukko-Pekka Luukkonen has a .944 save percentage to start the year. Maybe that’s more reassuring. No matter how many soft goals Oettinger lets in, how scrambly he looks, you still give him the chance. After all, who would you rather have in net?
I like that trade package for Tanev a lot actually. Maybe throwing in Dadanov helps balance the financials. I don't see him as horrendously overpaid so Calgary may be willing to try him out with Huberdeau.
Re. Harley and his ‘interview’…do the NFL and NBA conduct INTERVIEWS ? If so, how are we having to tolerate the loud-mouth buttholes that disrespect the Flag that represents the Great Country that gave/ gives them the opportunity to compete ? Give the players the talent and expertise to compete and let the Teams bring them along into the NHL. Every organization is going to have eggheads….Lord help us if we get another Lebron or Kapperstick. Sorry Jon….I don’t care for Taylor Swift’s music either.