2024-2025 Blues Multi-Purpose Thread.

[BlueDream]

Those Mike Reilly and Vince Dunn numbers are so absurd they sound made up. That’s insane.

 

[Linkens Mastery]

A lot of values for defensemen are weird for his model.
Here are some that raise my eyebrow.


Sergechev has a market Value of 1.3mil
Jake McCabe has a Value of 7.1mil
Matt Roy has a Value 9.7mil

 

[Reality Czech]

To put it in pragmatic perspective, Dunn is getting paid almost 1 MIL more than Parayko this year.

I'd rather have Parayko than Dunn. Seems like everyone is infatuated with defensemen who put up points and actually playing defense is secondary. Dom's model seems more suited for fantasy hockey than real hockey.

 

[Majorityof1]

I'd rather have Parayko than Dunn. Seems like everyone is infatuated with defensemen who put up points and actually playing defense is secondary. Dom's model seems more suited for fantasy hockey than real hockey.

I'd rather have both. You need both. Ideally you want your best D to do both. But if you cannot get it, there is a reason Dunn and Larsson are stapled at the hip at 5v5.

 

[STL fan in MN]

A lot of values for defensemen are weird for his model.
Here are some that raise my eyebrow.


Sergechev has a market Value of 1.3mil
Jake McCabe has a Value of 7.1mil
Matt Roy has a Value 9.7mil

If your model spits out results like this then you should probably tweak the model than publish the results…

 

[Reality Czech]

Was just listening to Hockey Sense with Strickland from a couple weeks ago when he interviewed Mathieu Joseph. Was a cool interview overall but I thought I'd pass this little tidbit along. When asked which Blues player he was looking forward to playing with, he mentioned how hard Leddy is to play against and how impressive his skating ability is. Not the answer I was expecting but clearly Leddy is respected by at least some of his peers.

 

[bleedblue1223]

I'll never understand Leddy's negative reputation that some seem to have of him.

 

[Reality Czech]

I'll never understand Leddy's negative reputation that some seem to have of him.

Just goes to show that even "average" NHLers are extremely skilled. I'll admit I was underwhelmed when we acquired him based on my limited knowledge, but he's been so much better than I expected. Such a solid pro.

 

[simon IC]

Just goes to show that even "average" NHLers are extremely skilled. I'll admit I was underwhelmed when we acquired him based on my limited knowledge, but he's been so much better than I expected. Such a solid pro.

I too was initially disappointed when I first heard about the Leddy acquisition. I am so happy I was wrong.

 

[bleedblue1223]

I remember when Strickland first mentioned Leddy as an option next to Parayko at that deadline and we were all pretty negative towards it. Part of me was optimistic because of his skating and puck-moving ability, and I think that type of player next to Parayko would really help him compared to guys like Mikkola or Scandella, but Leddy never faced the extreme deployment that he'd get next to Parayko. He exceeded my expectations in his current deployment.

Take out the disaster in Detroit and whatever happened in 17/18, and he's always been a solid player.

 

[Celtic Note]

I agree with most of what you said. I disagree with generalizing and projecting emotion onto people for pointing out the flaws in the model. That's strange to me.

Some people point are the flaws as a matter of fact. Others absolutely get bent out of shape. I am not saying all people critiquing the model are getting bent out of shape.

 

[Mike Liut]

I'll never understand Leddy's negative reputation that some seem to have of him.

Leddy is a quality partner for Parayko, but he’s not a true 1LD that we need to help this team go to the next level. Hopefully Broberg will be that player in another year, or we can trade for one.

 

[Stupendous Yappi]

Some people point are the flaws as a matter of fact. Others absolutely get bent out of shape. I am not saying all people critiquing the model are getting bent out of shape.

I don’t mind the model being flawed, i f it were presented with a little more humility, and less of the snarky comments about Blues fans. If youre going to present things like WE’RE the dummies for questioning the model, you deserve the flack.

I get that capturing statistics to reflect good defensive play is harder than counting shots and goals. There are probably some in-game events that could reflect impact better, but they’re not the available stats he can plug into a model. It would be cool to get to see the inner workings of some of the better proprietary analytics the clubs use.

 

[bleedblue1223]

And my main issue, the model might be pretty good for projected overall team results, I think it's reasonably close enough, and I don't follow different models enough to know where his would stack up against others. His issue is with individual assessments. When looking at overall team assessments, some of the issues on the individual level can be smoothed out and make less of an issue, but when you are going to do articles on the worst contracts in the league or best contracts in the league, those individual evaluations have to be dialed in, and his are not.

Now, he did start to make adjustments, Parayko being a specific player that he mentioned fueling that need, but they still aren't dialed in.

It would be interesting if he did a before and after and what went right/wrong at the end of the season. Take his projection from start of the season, how his model would evaluate the team based on actual stats, and compare that to how the team actually performed in the standings. Basically a comparison between the projection, the expected results, and the actual results. I'd hope that's something that happens behind the scenes, and would help readers that don't like him understand him and his model/methods better.

 

[Celtic Note]

I don’t mind the model being flawed, i f it were presented with a little more humility, and less of the snarky comments about Blues fans. If youre going to present things like WE’RE the dummies for questioning the model, you deserve the flack.

I get that capturing statistics to reflect good defensive play is harder than counting shots and goals. There are probably some in-game events that could reflect impact better, but they’re not the available stats he can plug into a model. It would be cool to get to see the inner workings of some of the better proprietary analytics the clubs use.

The Dom comments are just noise to me. He isn’t a fans of the Blues and to me he is just baiting people, so why take the bait? To each their own though.

I would love to see what clubs use too.

 

[taylord22]

My baseline: Dom's model is near perfect engagement bait. It's accurate enough, but also unabashedly ludicrous.

If it was anything close to being a real tool, NYT would've productized it. Therefore it remains a toy.

 

[BadgersandBlues]

The problem with Dom's model, as well as any sports analysis, is that you can't account for an event that -doesn't- occur. You can look at rates and all that, but it's pretty hard to figure out how much a guy stops anything from happening since.....it didn't happen. This is why Dom's model looks silly for defensive defenseman, and that's coming from someone who thinks Parayko isn't worth his contract either, for exactly the reasons Dom mentions. I don't think Parayko is a 2.5M dollar player, but I also don't think he's worth 6.5 either. I've touched on my thoughts on him in a different post last year, and I still stand by those comments.