Into The Weeds: Unstacking Micah Blake McCurdy's preview of the 2024-2025 Dallas Stars
Fancy models love the Stars. But to what end?
Why analytics are super important.
That could probably be the title to a lot of my articles here. I certainly hope that’s not the case, because that’s never been the point. Heck, it’s not even something I enjoy. I don’t enjoy coming across words like “covariates” and pretending like I can understand it. So why bother?
Two reasons. One is very simple, namely that I wouldn’t be doing my job if I didn’t make an honest attempt to understand a new language for hockey. Alison Lukan stands out in hockey and wins awards because she’s someone that makes that effort. She doesn’t assume she knows everything. In fact, she assumes she doesn’t know hockey at all — not because she’s a novice, or isn’t highly educated but because she respects that there’s always something to learn. Her article analyzing Brandon Montour’s skating with the help of assistant coach Jessica Campbell is a great example.
So that’s part of it; respecting what we don’t know. The other part is respecting the unique nature of hockey. I don’t know about you, but I love hockey. Why am I giving you this 2-out-of-10 description of an obviously shared interest? Because for me, part of that loves comes from its complexity.
Hockey is a game of different natures always interacting — quickly, and often violently — with one another. There is the specific nature of hockey; for example the properties of a great skating defenseman like Miro Heiskanen, and the tendency towards shutting down elite forwards by leveraging that specific nature. There is the local nature of hockey; that is, the properties of context, like the way four forwards on a power play create more offense than a PP with two defenders, or the impact of a faceoff. And then there are the universal laws of hockey: trailing teams will create more offense than leading teams, home teams have a slight advantage, special teams increases the odds of scoring, and nobody wants to listen to Pierre McGuire. This triple helix makes for a viewing experience unlike anything else in sports.
That’s a lot. And that’s why Micah Blake McCurdy’s season preview includes a lot; not just a lot of player data, and team data, but some expensive computers to simulate each game (literally a million times), and their theoretical outcome based on identifying the different natures of hockey interacting all at once to give us an estimate of what will happen. So…you ready?
The bottom line
I recommend reading Micah’s preview in full if you’re curious about the data.1 Based on Micah’s calculus, Dallas has a 24 percent chance to finish first in the division, every so slightly above the Colorado Avalanche, which would have them finishing fifth in the entire NHL.
Gut reaction: that’s how I see it too. Sort of. I think Minnesota and Nashville will duke it for that third spot. Maybe one of them does a Vancouver: finishing higher than they deserve from an inflated PDO bender, and never really coming down until the postseason. It’s interesting to see the Blues so low. I double checked, and realized why: they were top six in save percentage last season — the only team in that range to miss the postseason. It’s hard to believe they’ll get goaltending like that again.
But forget about the Blues. Let’s focus on (whose roster is set2) the Stars: from their even strength value to their discipline compared to the rest of the league, the numbers are in.
Programming note: I noticed quite a few new subscribers. Thank you! And just in time. I revamped my About Page for those curious how this little Stack that could, works. I’m excited for the new season, and I hope you are too.