Roster/Rumors/Speculation/Trade Talk - 2023-24: Hotel California

[Throttle]

The problem with crediting Dubas for "building" TOR is that they had 105 points the last year Lou was GM. So Dubas basically inherited Lou's team, signed Tavares, and improved the team by 1 point in the past 4 years with zero playoff success. I think they should name a trophy after him.

Pretty solid summary in two sentences of that situation. Well done.

 

[doublechili]

Sure, but since that time their top players all received massive raises. They have been able to maintain their regular season success while having to get cheaper at the fringes. There is value in that. Other GMs may have traded a Nylander, Rielly, etc. They have been able to keep the core intact while eliminating bad contracts.

The postseason results speak for themselves. They have not gotten past the first round. They lost to some good teams. It happens. Like, I said it isn't perfect. But the process is good, but needs tweaks.

However, this is an Isles board. And my greater point isn't about Dubas. I just find it overall interesting how anti-analytics this board is as a whole. Dubas in the GM embodiment of all things analytics. One can say he hasn't been good enough. And that is fair. I agree. But to say he sucks, is not accurate. There is value in maintaining a 100+ point team every season in a salary cap world. Some posters on here find zero to no value in the data.

Sakic and Yzerman are the standards in todays NHL. Two guys heavily invested in data.

I think some analytics has a place as a tool among other things. I think the question is whether it's elevated higher than it should be.

 

[CupHolders]

Sure, but since that time their top players all received massive raises. They have been able to maintain their regular season success while having to get cheaper at the fringes. There is value in that. Other GMs may have traded a Nylander, Rielly, etc. They have been able to keep the core intact while eliminating bad contracts.

The postseason results speak for themselves. They have not gotten past the first round. They lost to some good teams. It happens. Like, I said it isn't perfect. But the process is good, but needs tweaks.

However, this is an Isles board. And my greater point isn't about Dubas. I just find it overall interesting how anti-analytics this board is as a whole. Dubas in the GM embodiment of all things analytics. One can say he hasn't been good enough. And that is fair. I agree. But to say he sucks, is not accurate. There is value in maintaining a 100+ point team every season in a salary cap world. Some posters on here find zero to no value in the data.

Sakic and Yzerman are the standards in todays NHL. Two guys heavily invested in data.

I don’t think people are anti-analytics. I’m actually not myself. Even though some of my earlier exchanges portray me as such.

But I’m anti how pseudo professionals like Dom present it and HFBoard members misuse it to support their narrative agendas. Just to clarify, I AM NOT referring to you personally with that comment.

I am of the Stephen Valiquette school of thought when he stated something to the effect… “the numbers are not the tell all end of the story, they merely point to things that need to be looked at.”

What I have noticed with the media analyticians and their social media lemmings (cough, cough Rangers and Leaf fans). They do the following:

1A) Diminish that these are predictive metrics that focus on probability…

1B)…to the point of completely disregarding and discrediting actual outcomes.

I even wouldn’t have a problem with the justification that the actual results were merely outliers to their models. Except that many of the most ardent and belligerent defenders of analytics tend to be unwilling to accept the possible flaws in their CURRENT metrics and analysis.

So the same individuals who are imperiously arguing WAR stats and HDC% or whatever is the latest and greatest metric… were also arguing five years ago that CORSI was the best indicator of success over actual winning records.

Which is also fine, EXCEPT five years ago when there were outliers to CORSI, these individuals would just double-down instead of accepting that MAYBE the metrics at that time were simply too nascent and we should keep an open mind for possible improvements.

So analytics are fine, belligerent Dubasses using analytics to justify a narrative are not.

 

[danteipp]

I think some analytics has a place as a tool among other things. I think the question is whether it's elevated higher than it should be.

To me, analytics are best used to narrow down the field on individual players, who you will then devote more time towards scouting.

I use analytics for my fantasy baseball teams, primarily for selecting pitchers and it really helps later in the season as data builds, and to a lesser extent for fantasy football as well, mostly for RBs, WRs and TEs.

But applying it towards an entire team and projecting overall performance, where there are so many other things and outliers to consider like health, locker room culture, luck, even personal lives and traits, etc. is a tough call and a fools errand.

 

[Mike C]

To me, analytics are best used to narrow down the field on individual players, who you will then devote more time towards scouting.

I use analytics for my fantasy baseball teams, primarily for selecting pitchers and it really helps later in the season as data builds, and to a lesser extent for fantasy football as well, mostly for RBs, WRs and TEs.

But applying it towards an entire team and projecting overall performance, where there are so many other things and outliers to consider like health, locker room culture, luck, even personal lives and traits, etc. is a tough call and a fools errand.

Funny, I have never used analytics for fantasy BB. I'm an old timer. Throw the ball, hit the ball. Maybe I should enter into the new millennium?

 

[Mike C]

I think some analytics has a place as a tool among other things. I think the question is whether it's elevated higher than it should be.

EXACTLY!!

 

[Arthur Morgan]

Sorry if this is the wrong spot

Well.... Looks like I've become an NYI fan as long as Sorokin is part of this team. I just found away to get Sorokin for
Blackwood
Brannstrom
Nolan Allan
2023 1st

So whos the next in line goalie to backup Sorokin?
Varlamov going to be moved?

 

[Mike C]

Sorry if this is the wrong spot

Well.... Looks like I've become an NYI fan as long as Sorokin is part of this team. I just found away to get Sorokin for
Blackwood
Brannstrom
Nolan Allan
2023 1st

So whos the next in line goalie to backup Sorokin?
Varlamov going to be moved?

Lou has stated that Varlamov will be staying for whatever its worth

 

[Arthur Morgan]

Lou has stated that Varlamov will be staying for whatever its worth

ok cool well I dont wanna acquire Varlamov and his 5m contract how many games do u guys expect Sorokin to get this year? like 60 be nice but I can understand if Varlamov takes 30

 

[CupHolders]

Funny, I have never used analytics for fantasy BB. I'm an old timer. Throw the ball, hit the ball. Maybe I should enter into the new millennium?

No worries bro. Some of the latest research is showing that it ain't all that bad for you...

I kid, I kid!

 

[Mike C]

ok cool well I dont wanna acquire Varlamov and his 5m contract how many games do u guys expect Sorokin to get this year? like 60 be nice but I can understand if Varlamov takes 30

I would think sorokin for 50 maybe 55. The fear for me is that he'd wear down late if overworked early. A lot will probably depend on how the team starts out and where the playoff positioning looks. After last season, with this group, it's a playoff or bust mentality more than usual I suspect

No worries bro. Some of the latest research is showing that it ain't all that bad for you...

View attachment 574981


I kid, I kid!

LOVE IT!!

 

[CupHolders]

Sorry if this is the wrong spot

Well.... Looks like I've become an NYI fan as long as Sorokin is part of this team. I just found away to get Sorokin for
Blackwood
Brannstrom
Nolan Allan
2023 1st

So whos the next in line goalie to backup Sorokin?
Varlamov going to be moved?

I'm confused by this and your subsequent post... Are you writing of a fantasy trade where you acquired Sorokin for your team?

As for Varlamov, the earliest he likely get traded if at all is at the deadline. Isles like to platoon their goalies, so if everyone is healthy Sorokin likely comes in at 50 or so games.

 

[Arthur Morgan]

I'm confused by this and your subsequent post... Are you writing of a fantasy trade where you acquired Sorokin for your team?

As for Varlamov, the earliest he likely get traded if at all is at the deadline. Isles like to platoon their goalies, so if everyone is healthy Sorokin likely comes in at 50 or so games.

yeah in my 32 team dynasty league ok cool I figured he get like 50+ just wasnt sure if Varlamov was going to be moved or not.
Ill prob look to get whoever eventually backups Sorokin.
so Ill be cheering for NYI success
if Sorokin is playing the Leafs Ill prob cheer for Sorokin to give me the W lmao

 

[CupHolders]

yeah in my 32 team dynasty league ok cool I figured he get like 50+ just wasnt sure if Varlamov was going to be moved or not.
Ill prob look to get whoever eventually backups Sorokin.
so Ill be cheering for NYI success
if Sorokin is playing the Leafs Ill prob cheer for Sorokin to give me the W lmao

Just another heads up. Isles have NO goalie prospects ready to make the jump. So unless it's something catastrophic (long-term injuries), I wouldn't expect anyone to take time away from either.

If it is something catastrophic... well then you might want a backup goalie from an entirely different team.

 

[Arthur Morgan]

Just another heads up. Isles have NO goalie prospects ready to make the jump. So unless it's something catastrophic (long-term injuries), I wouldn't expect anyone to take time away from either.

If it is something catastrophic... well then you might want a backup goalie from an entirely different team.

I was thinking that too. still got Free Agency to go but looks like Jon Gillies is available not sure how active of a role he has I havent looked in a while and I think Craig Anderson is available too off the top of my head
I got Daws as well so if one of NJ's guys go down Im sure hes gnna get the call
I also wonder about Connor Ingram not sure who NSH has in store for backup but he could be an option

appreciate the advice btw

 

[danteipp]

Funny, I have never used analytics for fantasy BB. I'm an old timer. Throw the ball, hit the ball. Maybe I should enter into the new millennium?

Quick off topic note since you asked. For selecting pitchers, especially making moves during the season, the information sites like Fangraphs contains, such as tracking drop or velocity changes on pitches (you can see who added a new pitch, who is say throwing their curve harder or changed their fastball-changeup speeds), pitch value, expected ERA, etc. is invaluable. At least to me, I think it really helps in making decisions. Same for hitters. They give you data year-over-year on contact and hard hit rates, GB:FB and lots of other interesting tidbits. I have done better making calls on pitchers. Especially knowing when to cut bait due to declining peripherals.

 

[Mike C]

Quick off topic note since you asked. For selecting pitchers, especially making moves during the season, the information sites like Fangraphs contains, such as tracking drop or velocity changes on pitches (you can see who added a new pitch, who is say throwing their curve harder or changed their fastball-changeup speeds), pitch value, expected ERA, etc. is invaluable. At least to me, I think it really helps in making decisions. Same for hitters. They give you data year-over-year on contact and hard hit rates, GB:FB and lots of other interesting tidbits. I have done better making calls on pitchers. Especially knowing when to cut bait due to declining peripherals.

I'm sure my disdain for analytics has at least partially to do with it over saturates my feeble brain with information overload!

 

[YearlyLottery]

Funny, I have never used analytics for fantasy BB. I'm an old timer. Throw the ball, hit the ball. Maybe I should enter into the new millennium?

Eno Sarris is someone fantastic to follow for pitchers in fantasy baseball. Absolutely amazing.

 

[PK Cronin]

I do not mean to offend. I understood the crux of your post to be anti-analytics. If I misunderstood that, then I apologize.

I'm not offended, you're just not providing anything of substance that I haven't heard already and you're intentionally ignoring what I've brought up.

You seem to be set in your ways regarding analytics. It's a broad term and their are dozens of well known public models. Most teams also have private models they adhere to. Overall, the best way to determine their worth is their predictive value. I happen to like Dom's model for instance because I've used his model for betting purposes with great success. There are people I know who bet professionally, who use his model to make bets. They have their own models as well. Math is not a bad thing. I'm not really sure how you extrapolated the data of the above article, but by any mathematical principle, Dom's model is good.

I couldn't care less if you use it for gambling purposes. His model "works" in the regular season and fails in the playoffs. His success rate in the regular season isn't super impressive since it's not hard to predict which teams make the playoffs.

This brings me to Dubas. People hate on Dubas because he is very openly heavy on analytics. There's no other reason to hate him. There's also no reason to think he sucks.

There's plenty of reasons to dislike him.

His notable trade history:
- Trading down and drafting Rasmus Sandin
- Salary dumped Matt Martin to Isles for no cost
- Grundstrom, Durzi, 1st rounder (Bjornfoot) for Muzzin
- Salary dumped Marleau for 1st rounder (Seth Jarvis) - paid a lot to get rid of a bad Lou contract
- Traded Zaitsev and Brown to Senators in another salary dump (Zaitsev) - cost him Brown, who is a pretty good player - Zaitsev was another bad Lou signing
- Traded Kadri for Kerfoot and Barrie - this was actually a bad trade as I thought they'd be better off keeping Kadri at 4.5 mil rather than JT at 11+.
- Acquire Campbell and Clifford for basically nothing - Campbell was mostly a positive for them at a position Dubas never fully addressed over his tenure
- Traded Mason Marchment for Dennis Malgin- This was a bad trade as Marchment became pretty damn good
- Traded a 1st round pick for Nick Foligno as a rental - this was an overpayment as part of an over correction for not being "tough" enough. Foligno contributed virtually nothing during their run.
- Traded for Mark Giordano for a 2nd and 3rd round picks, then signed him to league minimum- IMO an excellent move
- Traded for Matt Murray - IMO a bad move as Murray is not very good

His trading history above was first getting rid of the bad contracts Lou collected. Then, he started dumping good players to make room for the JT signing. Which is a signing they did not need to make. The flat cap made that decision even worse. Dubas has also been unable to get an elite level goalie. The defense has also been an issue at times. Again, because of the top heavy contract structure they have.

You're just wrong. He made moves he didn't have to make in order to prioritize things he didn't need to prioritize. He grabbed Tavares, that f***ed their whole cap situation. Instead of dealing some of those forwards he doubled down on having a high powered offense and rarely/barely addressed the defense. Now the defense is meh and his goaltending has gotten worse. His team has had zero success in the playoffs and you're a fool if you think he's the reason that the team is a regular season juggernaut because of him.

His roster isn't perfect. There have always been questions about defense and goaltending. However, that is still a 105 point team on average since he took over despite Lou's best efforts to submarine the team with awful contracts.

Yeah, no.

He's the GM of a top 5 team in the league. They just lost to the Lightning in a 7 game series, where they were just as good during the series. That was an excellent series with two excellent teams. 4 years in, this is Dubas' team.

But we all have our biases. Right or wrong, mostly we stick with them. Teams that have ignored the data "analytics" provide have struggled. Teams that have embraced it, haven't. The proof is in the results. All you have to do is look at the standings and the recent cup winners. There are obvious outliers (Canadiens), but for the most part they are teams that squeeze the most value out of their cap space. Data is what allows them to do that. "Analytics" is just a polarizing word to describe data.

Which teams ignored analytics? Also, you mentioned teams use their own analytics, not Dom's public model, so his model isn't relevant.

I'm not against analytics, I'm against dumb models.

You've still conveniently ignored the question about how his model finally gave respect to the Islanders after undervaluing them for three straight years. Why and how does that happen when the team is exactly the same?

 

[JPIsles18]

I'm not offended, you're just not providing anything of substance that I haven't he

I'm not offended, you're just not providing anything of substance that I haven't heard already and you're intentionally ignoring what I've brought up.



I couldn't care less if you use it for gambling purposes. His model "works" in the regular season and fails in the playoffs. His success rate in the regular season isn't super impressive since it's not hard to predict which teams make the playoffs.



There's plenty of reasons to dislike him.



You're just wrong. He made moves he didn't have to make in order to prioritize things he didn't need to prioritize. He grabbed Tavares, that f***ed their whole cap situation. Instead of dealing some of those forwards he doubled down on having a high powered offense and rarely/barely addressed the defense. Now the defense is meh and his goaltending has gotten worse. His team has had zero success in the playoffs and you're a fool if you think he's the reason that the team is a regular season juggernaut because of him.



Yeah, no.


Which teams ignored analytics? Also, you mentioned teams use their own analytics, not Dom's public model, so his model isn't relevant.

I'm not against analytics, I'm against dumb models.

You've still conveniently ignored the question about how his model finally gave respect to the Islanders after undervaluing them for three straight years. Why and how does that happen when the team is exactly the same?

I agree with your Tavares point. I think he made a huge mistake prioritizing forwards over defense. I think that's what cost the team the first few years in the playoffs. He has been trying to rectify the situation ever since. The 2021-22 version of the Leafs was the best version yet. Mostly from internal improvement from the drafted defensemen and acquisition of a Giordano. They had the Lightning on the ropes. The series was about as equal as any during the playoffs. They lost. I'm glad they lost. However, there was no shame in losing to that Lightning team.

As far as Dom's model goes, he explains the change right here in his preview:

"To those that would listen, I often preached patience with the Islanders’ rating per this model. Many cited the Trotz effect and without any explicit coaching effects baked in, it would take time for that to settle in. That was especially true for a team whose focus on shot quality over everything wasn’t easily accounted for by public analytics.

But that also ignores that last year was a very different season for the Islanders compared to the first two under Trotz, where a bridge between eye test and analytics was finally formed. Islanders fans screamed as loud as they could that this team controlled quality to an elite degree, but it wasn’t until last season (or the 2020 playoffs rather) where that claim could be backed up by public analytics. In 2018-19, they were a solid 51.1 percent expected goals team, but that dropped to 49.9 percent the following year (with a below average actual goals rate). Then in the playoffs, the team finally turned it on and didn’t look back over the 2021 season. The Islanders had a 53.9 percent expected goals rate driven by elite defense last season, ranking top five in both categories. Before an injury to their captain, they were even higher."

Dom's model is just a derivative off of xGF. It has evolved away from something like CORSI/FENWICK, which just valued quantity over quality. Dom and other public models were very high on the Isles coming into this season because of the above points. The Isles became an elite defensive team with above average goaltending. Last year their defense took a huge hit, despite having improved goaltending from Sorokin. The reasons for that can be debated ad nauseam (COVID, road trip, taxing style over 82 games, Chara and Green in the top 6, too many middle 6 forwards, etc).

 

[PK Cronin]

I agree with your Tavares point. I think he made a huge mistake prioritizing forwards over defense. I think that's what cost the team the first few years in the playoffs. He has been trying to rectify the situation ever since. The 2021-22 version of the Leafs was the best version yet. Mostly from internal improvement from the drafted defensemen and acquisition of a Giordano. They had the Lightning on the ropes. The series was about as equal as any during the playoffs. They lost. I'm glad they lost. However, there was no shame in losing to that Lightning team.

As far as Dom's model goes, he explains the change right here in his preview:

"To those that would listen, I often preached patience with the Islanders’ rating per this model. Many cited the Trotz effect and without any explicit coaching effects baked in, it would take time for that to settle in. That was especially true for a team whose focus on shot quality over everything wasn’t easily accounted for by public analytics.

But that also ignores that last year was a very different season for the Islanders compared to the first two under Trotz, where a bridge between eye test and analytics was finally formed. Islanders fans screamed as loud as they could that this team controlled quality to an elite degree, but it wasn’t until last season (or the 2020 playoffs rather) where that claim could be backed up by public analytics. In 2018-19, they were a solid 51.1 percent expected goals team, but that dropped to 49.9 percent the following year (with a below average actual goals rate). Then in the playoffs, the team finally turned it on and didn’t look back over the 2021 season. The Islanders had a 53.9 percent expected goals rate driven by elite defense last season, ranking top five in both categories. Before an injury to their captain, they were even higher."

Dom's model is just a derivative off of xGF. It has evolved away from something like CORSI/FENWICK, which just valued quantity over quality. Dom and other public models were very high on the Isles coming into this season because of the above points. The Isles became an elite defensive team with above average goaltending. Last year their defense took a huge hit, despite having improved goaltending from Sorokin. The reasons for that can be debated ad nauseam (COVID, road trip, taxing style over 82 games, Chara and Green in the top 6, too many middle 6 forwards, etc).

So it's exactly what I said originally, he's continually adjusting the model as he goes which means that using previous model success as a predicator of future success is silly. It'd be one thing if he created models using historically great teams and said they all have [x] in common and it's constant over every era of hockey, but that's not the case. Models also accept trends in the game and would punish anything outside of those trends. Think of the CORSI/FENWICK discussion from years past. Using those as proxies for possession and/or goal scoring predictability in a historic sense is fine, if you're doing what I mentioned above with historic teams. What becomes a problem is once teams and players are aware of the usage of that stat and they believe it correlates to more wins the more you'll see them adjust strategy to fit it into their game plan. For example, if Trotz bought into CORSI/FENWICK he could emphasize quantity over quality to help inflate those statistics even though quality over quantity was working just fine for his team.

There are just a lot of issues I have with all of these things.

The fact that the eye test told me what the model couldn't for two years is also an issue I have and helps demonstrate that the models are still behind a person's ability to understand what's happening in real time.

 

[Chardo]

I just don't see the reliability of analytics for a team sport played on ice, with so many factors out of a player's control.

 

[SI]

With regards to a Josh Bailey trade...

I have two options I wanted to toss out there.

#1. I wonder if a Josh Bailey (2y/5AAV) for a John Moore (1y/2.75) deal could work - Anaheim does have cap space and need of a vet F like Bailey. John Moore does not seem to feature on the roster. Ducks could possibly flip Bailey at the deadline for picks. The 2.25 in savings and potentially more if they bury Moore (2.65), this could be a way to create space for Kadri.
EDIT - The Isles would still be slightly over the cap, so the signings of Dobson, Kadri, Bellows, and Romanov would have to be less than expected.

#2. I wonder if a Josh Bailey for Ben Bishop (1y/4.95) deal could work too? Isles would move Bishop to LTIR and that would create space for Kadri. I don't love the idea of having a player on LTIR, b/c they cannot accrue cap space throughout the season, but maybe they end trading Bishop later to get them out of it. Below is a screenshot of what it could look like.

 

[SI]

 

[PK Cronin]

I just don't see the reliability of analytics for a team sport played on ice, with so many factors out of a player's control.

The further the statistic gets from the actual thing it's supposed to judge the more useless it becomes.

i.e. Possession leads to more goals, which leads to more wins, but we can't actually measure possession so we'll use shot attempts instead.

Now that the NHL has trackers in the equipment, I suspect the NHL will have access to much more useful things than we'll see (unless they make it public). They can track actual possession, they can track where the puck goes, velocity of the puck, how fast the players are moving, etc. and it will be far better than what we're seeing now.