Confirmed with Link: PLD 1 year 6 Million(accepted qualifying offer)

[hockeyarena]

I wonder if there is a lack of accountability with the jets?

Pacioretty believes Golden Knights organization lacks accountability

When his request to be moved out of Montreal was granted by a trade to the Vegas Golden Knights, there was some relief for Pacioretty. What he didn't expect was how vastly different the environment was, especially for a team that was contending right from its first season as an expansion club.

www.sportsnet.ca

Scheifele didn't try the entire year and Laine played video games for 12 hours straight. So ya.

 

[nobody imp0rtant]

Scheifele didn't try the entire year and Laine played video games for 12 hours straight. So ya.

What Laine did on his own time doesn't concern me. What ScheifMe does on the team's time, does.

 

[Imcanadianeh]

Scheifele didn't try the entire year and Laine played video games for 12 hours straight. So ya.

Well apparently when the leadership core of the Jets (Wheeler, Scheifele, Buff) wanted more from Laine he disapproved and wasn’t happy about it, now lots of people blame Wheeler for that and seems to be where a lot of the Wheeler hate comes from.

“I think he chafed under some of the leadership there. The guys at the top of that food chain are hard driving guys. They expect you to buy into the program and I think that they felt he didn’t buy in enough and I think he felt that some of the things that they wanted were ridiculous.” - Elliotte Friedman

A double edged sword for Wheeler, some people will say he’s an Asshole for trying to get players to buy in and play to their potential or he’s a bad leader for not getting players to buy in.

 

[KingBogo]

Well apparently when the leadership core of the Jets (Wheeler, Scheifele, Buff) wanted more from Laine he disapproved and wasn’t happy about it, now lots of people blame Wheeler for that and seems to be where a lot of the Wheeler hate comes from.

“I think he chafed under some of the leadership there. The guys at the top of that food chain are hard driving guys. They expect you to buy into the program and I think that they felt he didn’t buy in enough and I think he felt that some of the things that they wanted were ridiculous.” - Elliotte Friedman

A double edged sword for Wheeler, some people will say he’s an Asshole for trying to get players to buy in and play to their potential or he’s a bad leader for not getting players to buy in.

And did Wheeler push any harder than Captain Serious did with the Hawks Cup runs or were there just a different makeup of players behind him that didn’t want it quite so much?

 

[surixon]

Well apparently when the leadership core of the Jets (Wheeler, Scheifele, Buff) wanted more from Laine he disapproved and wasn’t happy about it, now lots of people blame Wheeler for that and seems to be where a lot of the Wheeler hate comes from.

“I think he chafed under some of the leadership there. The guys at the top of that food chain are hard driving guys. They expect you to buy into the program and I think that they felt he didn’t buy in enough and I think he felt that some of the things that they wanted were ridiculous.” - Elliotte Friedman

A double edged sword for Wheeler, some people will say he’s an Asshole for trying to get players to buy in and play to their potential or he’s a bad leader for not getting players to buy in.

If I recall it was more around his off ice habits that they had an issue with. There was something out about him wanting to sit in his hotel room and play video games in his down time instead if hanging out. They also clashed about how he didn't like to think about hockey once out of the rink whereas Blake and Mark are hockey 100% of the time.

I think it was clearly a personality clash. I think both parties are at fault in this. Laine needed to train better in the offseasons and Mark and Blake should have only been concerned with his on ice and in room contributions. What he does in his down time is none of their damn business.

 

[Imcanadianeh]

If I recall it was more around his off ice habits that they had an issue with. There was something out about him wanting to sit in his hotel room and play video games in his down time instead if hanging out. They also clashed about how he didn't like to think about hockey once out of the rink whereas Blake and Mark are hockey 100% of the time.

I think it was clearly a personality clash. I think both parties are at fault in this. Laine needed to train better in the offseasons and Mark and Blake should have only been concerned with his on ice and in room contributions. What he does in his down time is none of their damn business.

Who really knows what all happened, even Friedman who probably knows and hears a lot more than us uses the words “I think”.

But it was pretty well documented that Laine really liked to play fortnite and he struggled mightily for much of that season (18-19) in the last 60 games of the season he had just five 5v5 goals and 11 goals total, he had the worst GF% and xGF% of all forwards during that stretch, so I could see why teammates weren’t happy with him playing video games possibly all night when he was arguably the worst forward on the team for much of that season.

I could also see why teammates were frustrated with Laine because it sounds like he was fine with isolating himself away from the team.

 

[tbcwpg]

I just saw an article the other day listing Suzuki as one of the top ten worst contracts in the NHL, ranked at #9 with Ben Chiarot at #10 for comparison (Tyler Seguin, Seth Jones and more: NHL's 10 worst contracts, 2022 edition). Yet I keep seeing his name come up as someone we want in a trade for PLD. Why is that? Are people banking on these models being wrong?

The guy who made the model says in that article he thinks the model will be wrong on Suzuki. It also had Morrissey on there last season and now he's graded at a B level for contract value. Things change from year to year.

 

[SCP Guy]

Who really knows what all happened, even Friedman who probably knows and hears a lot more than us uses the words “I think”.

But it was pretty well documented that Laine really liked to play fortnite and he struggled mightily for much of that season (18-19) in the last 60 games of the season he had just five 5v5 goals and 11 goals total, he had the worst GF% and xGF% of all forwards during that stretch, so I could see why teammates weren’t happy with him playing video games possibly all night when he was arguably the worst forward on the team for much of that season.

I could also see why teammates were frustrated with Laine because it sounds like he was fine with isolating himself away from the team.

We should embrace the hermit hockey players in the NHL… they prob don’t care about all the attractions bigger cities have over Winnipeg and are happy to sit in their condos, order skip and enjoy living the life lol

 

[surixon]

Who really knows what all happened, even Friedman who probably knows and hears a lot more than us uses the words “I think”.

But it was pretty well documented that Laine really liked to play fortnite and he struggled mightily for much of that season (18-19) in the last 60 games of the season he had just five 5v5 goals and 11 goals total, he had the worst GF% and xGF% of all forwards during that stretch, so I could see why teammates weren’t happy with him playing video games possibly all night when he was arguably the worst forward on the team for much of that season.

I could also see why teammates were frustrated with Laine because it sounds like he was fine with isolating himself away from the team.

That poor back stretch happened after he torched November with a near record goal month.

 

[Buffdog]

The guy who made the model says in that article he thinks the model will be wrong on Suzuki. It also had Morrissey on there last season and now he's graded at a B level for contract value. Things change from year to year.

To be fair, when the Suzuki example happens, maybe you should toss the whole model. You can't have a "good" model and then have an outlier like that. That's proof of an obvious flaw

OR

Do what some of us have been saying for years... use advanced stats and models as tools to draw a conclusion, but don't label their outcomes as conclusions in and of themselves

 

[tbcwpg]

To be fair, when the Suzuki example happens, maybe you should toss the whole model. You can't have a "good" model and then have an outlier like that. That's proof of an obvious flaw

OR

Do what some of us have been saying for years... use advanced stats and models as tools to draw a conclusion, but don't label their outcomes as conclusions in and of themselves

It's not a flaw of the model necessarily. It's saying he's not worth close to $8m right now, but many of his comparables show he should grow into that in a couple of years.

As in, it's doing exactly what you say in your 2nd paragraph.

 

[garret9]

To be fair, when the Suzuki example happens, maybe you should toss the whole model. You can't have a "good" model and then have an outlier like that. That's proof of an obvious flaw

OR

Do what some of us have been saying for years... use advanced stats and models as tools to draw a conclusion, but don't label their outcomes as conclusions in and of themselves

1) Wrong. Don't throw out the baby with the water, and outliers will ALWAYS exist.

Now it would be very remarkable if any system existing in the real world could be exactly represented by any simple model. However, cunningly chosen parsimonious models often do provide remarkably useful approximations. For example, the law PV = nRT relating pressure P, volume V and temperature T of an "ideal" gas via a constant R is not exactly true for any real gas, but it frequently provides a useful approximation and furthermore its structure is informative since it springs from a physical view of the behavior of gas molecules. For such a model there is no need to ask the question "Is the model true?". If "truth" is to be the "whole truth" the answer must be "No". The only question of interest is "Is the model illuminating and useful?".​

2) How you should use analytics... close but not quite right.

You use a more Bayesian style thinking with priors and data. There's reasons why the model suggested these things. For example: with Morrissey it was because it doesn't have a "did your father pass away" input.

With Suzuki, it's still meaningful that he comes out as not a great bet to reach those numbers, that people who have performed statistically like him historically do not perform to that AAV.

It doesn't mean he will fail, but it's still meaningful, and you should adjust your prior beliefs, at least somewhat, because of it. How much you adjust will be based on the strength of the new evidence.

I do think Suzuki has enough upside that he can be really good... but look at two entirely other models... I'd be concerned about the fact that he has a high GAR (which often correlates with F contracts highly) and a very much less impressive xGAR (which is more predictive of how good a player will be in the future and likely then is better at evaluating true talent).

Again, not saying Suzuki won't end up a stud or at least good. But it's something that he crashed in 2way numbers as soon as he didn't have someone taking all the tough match ups for him AND that his underlying numbers have some red flags.

 

[snowkiddin]

It's not a flaw of the model necessarily. It's saying he's not worth close to $8m right now, but many of his comparables show he should grow into that in a couple of years.

As in, it's doing exactly what you say in your 2nd paragraph.

So is this list of Dom’s “the top 10 worst contracts in the NHL” or the “top 10 players who performed most poorly relative to what they were paid last season”? Because those are two different things.

Genuinely asking, as I’m unfamiliar with this article.

 

[snowkiddin]

Who really knows what all happened, even Friedman who probably knows and hears a lot more than us uses the words “I think”.

But it was pretty well documented that Laine really liked to play fortnite and he struggled mightily for much of that season (18-19) in the last 60 games of the season he had just five 5v5 goals and 11 goals total, he had the worst GF% and xGF% of all forwards during that stretch, so I could see why teammates weren’t happy with him playing video games possibly all night when he was arguably the worst forward on the team for much of that season.

I could also see why teammates were frustrated with Laine because it sounds like he was fine with isolating himself away from the team.

Is there any source that suggested he played Fortnite “possibly all night”? From what I recall, he enjoyed playing video games, and he played them enough that it was well-known (pretty sure players on other teams had PlayStation-related chirps).

 

[garret9]

So is this list of Dom’s “the top 10 worst contracts in the NHL” or the “top 10 players who performed most poorly relative to what they were paid last season”? Because those are two different things.

Genuinely asking, as I’m unfamiliar with this article.

This is his general process:

1) He has a model that evaluates player value (Game Score) with how many goals a player impacts the team goal differential. It's a solid model, although IMO it underrates defensive value (good or bad) and overrates scoring (which can be heavily usage dependent).

2) He then estimates how many goals equate to a win (see Pythagorean Win Expectations if you are curious in how people do this).

3) He looks at how the market has historically paid players per win they generate.

4) He looks at the closest historical statistical comparisons (in Game Score) to project a players future output over the durration of the remaining contract.

5) Then it's just simple subtraction (expected wins/dollar - market wins/dollar).

 

[snowkiddin]

This is his general process:

1) He has a model that evaluates player value (Game Score) with how many goals a player impacts the team goal differential. It's a solid model, although IMO it underrates defensive value (good or bad) and overrates scoring (which can be heavily usage dependent).

2) He then estimates how many goals equate to a win (see Pythagorean Win Expectations if you are curious in how people do this).

3) He looks at how the market has historically paid players per win they generate.

4) He looks at the closest historical statistical comparisons (in Game Score) to project a players future output over the durration of the remaining contract.

5) Then it's just simple subtraction (expected wins/dollar - market wins/dollar).

Thanks for this. For part three, does he factor in the rising cap? Is it how much they were paid relative to the cap at the time?

 

[tbcwpg]

So is this list of Dom’s “the top 10 worst contracts in the NHL” or the “top 10 players who performed most poorly relative to what they were paid last season”? Because those are two different things.

Genuinely asking, as I’m unfamiliar with this article.

So, essentially his model relates salary to the players impact in various categories (the basics like goals and assists but also chance creation etc etc), and then compares what they're making to what they're actually producing, and whether they're trending upwards or downwards. This article was a list of the 10 players who had rhe widest gap between their impact on each game and how much they were being paid. There wasn't any kind of "I think these are bad because..." but moreso just by the numbers and why the model says they're poor contracts.

His model, compared to several of the other ones floating around, is usually towards the top end of being correct. In Suzuki's case, it's partially a function of Montreal giving him that deal hoping he'll grow into it but not yet being an $8m player and also partially Montreal was pretty bad last year in everything and it dragged everyone on that team down.

 

[garret9]

Thanks for this. For part three, does he factor in the rising cap? Is it how much they were paid relative to the cap at the time?

I *believe* it's via cap % not $ to equate for interest, but not certain of Dom's method there.

 

[Imcanadianeh]

Is there any source that suggested he played Fortnite “possibly all night”? From what I recall, he enjoyed playing video games, and he played them enough that it was well-known (pretty sure players on other teams had PlayStation-related chirps).

His fortnite games were tracked pretty heavily back in the 18-19 season, eventually around March he stopped using his normal account because he was getting scrutinized for how much he was playing and the hours he played.

 

[Jets 31]

Well apparently when the leadership core of the Jets (Wheeler, Scheifele, Buff) wanted more from Laine he disapproved and wasn’t happy about it, now lots of people blame Wheeler for that and seems to be where a lot of the Wheeler hate comes from.

“I think he chafed under some of the leadership there. The guys at the top of that food chain are hard driving guys. They expect you to buy into the program and I think that they felt he didn’t buy in enough and I think he felt that some of the things that they wanted were ridiculous.” - Elliotte Friedman

A double edged sword for Wheeler, some people will say he’s an Asshole for trying to get players to buy in and play to their potential or he’s a bad leader for not getting players to buy in.

If Wheeler pushed the guys hard i have absolutely no problem with that, my issue with Wheeler is that for the last couple of seasons Ehlers should have been higher than him in the depth chart. Maurice was too chummy with Wheeler to see that Ehlers deserved to be higher on the depth chart and definitely should have been on the 1st PP unit. Again more of a problem with Maurice than Wheeler.

 

[ps241]

Interesting discussion. I always thought Phillip Danault was underrated in Montreal and its not surprising to me that Suzuki’s numbers suffered when he had to pick up some of that hard match up load from Danault after he was dealt to LA. The important thing to remember is the model isn’t saying Nick is a bad player, becasue Nick is a good player, its just saying he is overpriced currently.

 

[Bob E]

Full tin foil hat moment.

I would hope Chevy gets a return of Dach + Barron/Guhle + Florida’s 2023 1st in a deal with Montreal for PLD. If the Jets need to add something like Heinola to get it done, I’m ok with that.

I have a feeling Florida will take a step back this year and not make the playoffs in a tough division (call it the Maurice effect), so that pick might be top 15.

I also think the Jets stumble a bit, as they have gotten all they can from guys like Wheeler, Schmidt and Dillon, and the bottom 6 is still a mess as of now. They too could be looking at a top 10-15 pick.

With two picks in top 15, in a strong draft might be the rebuild kick start this team ultimately needs. Imagine being able to draft a C like Dvorsky and winger/C like Stramel. In a couple years…

Connor-Dach-Lambert (yes, I think he develops into a top 6 winger)
Perfetti-Dvorsky-Ehlers
McGroarty-Lucius-Stramel
Rachevsky-Gustafson-Chibrikov

Now that would be mostly scoring lines and there’s no Lowry or Appleton which means no chance this happens, but if these prospects work out, the Jets have an abundance of riches imo.

 

[Buffdog]

1) Wrong. Don't throw out the baby with the water, and outliers will ALWAYS exist.

Now it would be very remarkable if any system existing in the real world could be exactly represented by any simple model. However, cunningly chosen parsimonious models often do provide remarkably useful approximations. For example, the law PV = nRT relating pressure P, volume V and temperature T of an "ideal" gas via a constant R is not exactly true for any real gas, but it frequently provides a useful approximation and furthermore its structure is informative since it springs from a physical view of the behavior of gas molecules. For such a model there is no need to ask the question "Is the model true?". If "truth" is to be the "whole truth" the answer must be "No". The only question of interest is "Is the model illuminating and useful?".​

2) How you should use analytics... close but not quite right.

You use a more Bayesian style thinking with priors and data. There's reasons why the model suggested these things. For example: with Morrissey it was because it doesn't have a "did your father pass away" input.

With Suzuki, it's still meaningful that he comes out as not a great bet to reach those numbers, that people who have performed statistically like him historically do not perform to that AAV.

It doesn't mean he will fail, but it's still meaningful, and you should adjust your prior beliefs, at least somewhat, because of it. How much you adjust will be based on the strength of the new evidence.

View attachment 575097

I do think Suzuki has enough upside that he can be really good... but look at two entirely other models... I'd be concerned about the fact that he has a high GAR (which often correlates with F contracts highly) and a very much less impressive xGAR (which is more predictive of how good a player will be in the future and likely then is better at evaluating true talent).

Again, not saying Suzuki won't end up a stud or at least good. But it's something that he crashed in 2way numbers as soon as he didn't have someone taking all the tough match ups for him AND that his underlying numbers have some red flags.

You missed my point. What I'm saying is that a model can be used as a tool to assess something, but it's outputs should not be accepted as conclusions unto themselves (ie Player A is good because he has an xGF% or Player B is better than Player C because he has a better WAR). I'd wager that professional analytic types like yourself wouldn't say something like that, but the casual stats lovers do it all the time.

I appreciate you bringing a scientific equation into the convo. With the scientific method, you create a hypothesis, run the experiment, and then gather the data and assess. One of the biggest problems we see in published studies for all fields is data massaging and P hacking if the results of the study don't support the hypothesis. There seems to be an element of this here, where a guy makes a model that whiffs on Suzuki and then there's a bunch of "well, yeah but..." coming from the analytics crowd. Sure, that could be the case. OR the model isn't either as good as the discoverer liked, or isn't any good at all.

I appreciate the efforts to come up with objective, accurate and reliable methods to assess a player's impact. But this isn't baseball where you can take a career .275 hitter from one lineup and insert him in another lineup and over time he'll still be a .275 hitter (mostly because at it's core, baseball is more of an individual game than a team game - of course it matters where that hitter is in the lineup and who hits before and after him in the order - but it won't have as much of an impact as, say, putting Tanner Glass on Mark Sciefele's wing).

Again, I'm not anti-analytics. I know there's a place for them in hockey and I appreciate them. I also understand their limitations based on the sport they're trying to quantify.

 

[garret9]

This is his general process:

1) He has a model that evaluates player value (Game Score) with how many goals a player impacts the team goal differential. It's a solid model, although IMO it underrates defensive value (good or bad) and overrates scoring (which can be heavily usage dependent).

2) He then estimates how many goals equate to a win (see Pythagorean Win Expectations if you are curious in how people do this).

3) He looks at how the market has historically paid players per win they generate.

4) He looks at the closest historical statistical comparisons (in Game Score) to project a players future output over the durration of the remaining contract.

5) Then it's just simple subtraction (expected wins/dollar - market wins/dollar).

I forgot to add that every bad contract will look less bad each year, since he grades them by aggregate... and when you have one less year you look less bad.

 

[tbcwpg]

You missed my point. What I'm saying is that a model can be used as a tool to assess something, but it's outputs should not be accepted as conclusions unto themselves (ie Player A is good because he has an xGF% or Player B is better than Player C because he has a better WAR). I'd wager that professional analytic types like yourself wouldn't say something like that, but the casual stats lovers do it all the time.

I appreciate you bringing a scientific equation into the convo. With the scientific method, you create a hypothesis, run the experiment, and then gather the data and assess. One of the biggest problems we see in published studies for all fields is data massaging and P hacking if the results of the study don't support the hypothesis. There seems to be an element of this here, where a guy makes a model that whiffs on Suzuki and then there's a bunch of "well, yeah but..." coming from the analytics crowd. Sure, that could be the case. OR the model isn't either as good as the discoverer liked, or isn't any good at all.

I appreciate the efforts to come up with objective, accurate and reliable methods to assess a player's impact. But this isn't baseball where you can take a career .275 hitter from one lineup and insert him in another lineup and over time he'll still be a .275 hitter (mostly because at it's core, baseball is more of an individual game than a team game - of course it matters where that hitter is in the lineup and who hits before and after him in the order - but it won't have as much of an impact as, say, putting Tanner Glass on Mark Sciefele's wing).

Again, I'm not anti-analytics. I know there's a place for them in hockey and I appreciate them. I also understand their limitations based on the sport they're trying to quantify.

I understand what you're trying to say, but there's really a difference between a guy like Dom who developed this model and writes about it for a sports publication and someone like Garrett who's using it in the game.