Roster Building thread - Part X - (TDL edition)

[mas0764]

Would you sign him for 2 years $16M if he was a FA this off season? If the answer is no, then why is anyone trading for him without sending a bad contract back? He has a NTC so it's not like a team can trade for him if he wasn't interested in signing there as a FA.

Because the term is so low, yeah, I could see a team biting on that. Again, he's been terrible, sub-third pair defender, but many teams are dummies and pay for size. We are one of them.

Because not every team has a cap issue or the same roster as we do. He could have way more value for a team that is desperate for a physical presence on the blue line.

He would be just as bad anywhere, but other teams don't know that because they aren't smart and do the same old timey hockey things that all the other GMs do.

 

[hardnosed]

I guess I’m failing to see how things line up. This board is insistent Trouba is a bad defenseman. Not that he’s not worth his money. That he should be third pair or a healthy scratch. Yet at the same time someone is going to trade him, fake the full cap hit, and open up 8M space to be used elsewhere? Yea the real money is less that’s a minor factor. If the rangers retained 2M and then took someone like Greenway now they only have 3M to spend on his replacement which will be somehow who likely isn’t any good.

That's just the point. We have his replacement already in the lineup. Promote Schneider. Then you just need a #6 which you can get for 3 mil or less.

 

[NYR Viper]

Here's a scenario:

To NYR:
Cody Glass + TB 2nd '24 + COL 3rd '25 + EDM 4th '24

To NSH:
Jacob Trouba

Followed by:

To NYR:
Nico Sturm

To SJ:
Cody Glass


The Rangers buyout Goodrow and play Sturm as the 4C for next year, flanked by Edstrom and Vesey with Rempe sprinkled in.

The Rangers save $6m against the cap and recoup some picks.

 

[SA16]

He has not been sub-par third pairing level. You're just not properly understanding context.

 

[CLW]

Nicky Fotiu used to ride the subway with an axe in his hockey bag. Hahaha.
I started riding the subways in the mid 70’s. I spit out my coffee when I hear people talking about how it’s so much worse now than the good old days. Hahaha. Parts of NYC were like Beirut in the 70’s and 80’s. Yeah crime and such are cyclical and on the rise EVERYWHERE, but NYC is Disneyland now in comparison. There’s more than 8 million people, and more than a billion subway rides a year. Open 24 hours a day. With like 5 murders last year. That’s AMAZINGLY low.

Thar should have been Kakko's reply to the stupid NYC question: "I've only been here a minute, but I brought my axe..."

 

[mas0764]

He has not been sub-par third pairing level. You're just not properly understanding context.

Maybe you aren't.

We can debate exactly how much hyperbole that is, if any, but it's besides the point.

He's bad by any measurement.

Can we swindle another team into taking him? I think.

Getting him and his $8m hit off our books is only good for us no matter how you measure it.

 

[McRanger92]

Funny hearing Brian Boyle talk about how highly regarded Goodrow is and how bubble teams around the league would love to have a guy who can play within a defensive structure like him. He must be an idiot too.

Bodes well for us if we want to trade him though

 

[Pawnee Rangers]

Funny hearing Brian Boyle talk about how highly regarded Goodrow is and how bubble teams around the league would love to have a guy who can play within a defensive structure like him. He must be an idiot too.

Bodes well for us if we want to trade him though

If your choice is believing a former player/coach/trainer or some rando on HF boards. Always side with the rando. They have charts.

 

[McRanger92]

If your choice is believing a former player/coach/trainer or some rando on HF boards. Always side with the rando. They have charts.

Goodrow isnt very good but he serves a purpose that benefits the team over time.. Same with Trouba. The chart stuff has jumped the shark so much because too many zealots think its the only thing that matters.

 

[Pawnee Rangers]

Goodrow isnt very good but he serves a purpose that benefits the team over time.. Same with Trouba. The chart stuff has jumped the shark so much because too many zealots think its the only thing that matters.

If you were to ask players and coaches their opinions on certain players, it's probably going to be wildly different that what we read here. Like Boyle with Goodrow. They value things differently than fans do.

 

[Fitzy]

If your choice is believing a former player/coach/trainer or some rando on HF boards. Always side with the rando. They have charts.

True. In theory this isn't between any rando here and Boyle, though. It's between a few of the people doing the statistical work behind the scenes vs Boyle.

So it's more like the question- who would you rather have access to the nuclear launch codes: an eccentric rational choice theorist, or a former general?

 

[Pawnee Rangers]

True. In theory this isn't between any rando here and Boyle, though. It's between a few of the people doing the statistical work behind the scenes vs Boyle.

So it's more like the question- who would you rather have access to the nuclear launch codes: an eccentric rational choice theorist, or a former general?

I wasn’t calling out anyone here. But I’d rather listen to the guy that’s been there and has conversations with guys plugged in. Look, the charts say what they say about guys like Goodrow or Trouba, if that’s how people want to enjoy the game, good on them. But it’s not the end all be all nor does it define a players worth to his or her team. And Goodrow and Trouba are important whether guys here want to believe it or not.

 

[McRanger92]

True. In theory this isn't between any rando here and Boyle, though. It's between a few of the people doing the statistical work behind the scenes vs Boyle.

So it's more like the question- who would you rather have access to the nuclear launch codes: an eccentric rational choice theorist, or a former general?

Why didnt Tampa or the Rangers behind the scenes people stop them from giving up a first round pick or large contract? 2 pretty smart and successful organizations obviously loved him enough.

I just can't reconcile that the publicly available data is so good when it its seemingly constantly ignored by people in the industry. It's not like teams dont have their own proprietary data. Im really not anti-analytics but there are very few sites that can blend the underlying numbers with what's actually happening out there. Vally/CSA do it best imo and its because they do it "manually" and are the most cognizant of context. Moneypuck on the other hand is a joke because its just probability based. At least thats how they present themselves.

 

[Pawnee Rangers]

His model is weird. That model is not very good. This one looks bad because the team was defending a lead late and this model hates that. Don’s model does this. Her model does that. It’s just people looking for relevance in a crowded field of analysts. Seems pretty fugazi.

 

[Fitzy]

I wasn’t calling out anyone here. But I’d rather listen to the guy that’s been there and has conversations with guys plugged in. Look, the charts say what they say about guys like Goodrow or Trouba, if that’s how people want to enjoy the game, good on them. But it’s not the end all be all nor doesn’t it define a players worth to his or her team.

I completely agree with your take.

I just think there are insights to be gleaned from both.

I give the metrics folks no short of shit around here for their lack of epistemological modesty. But the metrics do keep getting better and better. We've come a long way from Corsi.

Why didnt Tampa or the Rangers behind the scenes people stop them from giving up a first round pick or large contract? 2 pretty smart and successful organizations obviously loved him enough.

I just can't reconcile that the publicly available data is so good when it its seemingly constantly ignored by people in the industry. It's not like teams dont have their own proprietary data. Im really not anti-analytics but there are very few sites that can blend the underlying numbers with what's actually happening out there. Vally/CSA do it best imo and its because they do it "manually" and are the most cognizant of context. Moneypuck on the other hand is a joke because its just probability based. At least thats how they present themselves.

You're not wrong. These are businesses with incentives. If metrics do turn out to be invaluable, we would expect to see them become a bigger and bigger part of the industry (And heck, teams could easily pay a few BAs reasonable salaries to do advisory research)

That said, the credibility of new methods tends to lag behind the acceptance and implementation. The innovators have to prove themselves, the traditionalists do not.

That said, everything I just said assumes there is important information to glean from metrics, which I genuinely believe there is. I just think the current crop needs to acknowledge the limitations of their research like any other field is supposed to.

 

[Pawnee Rangers]

I completely agree with your take.

I just think there are insights to be gleaned from both.

I give the metrics folks no short of shit around here for their lack of epistemological modesty. But the metrics do keep getting better and better. We've come a long way from Corsi.

I think they’re useful as well to an extent. Some help paint a fuller picture, like Valiquettes. I love seeing his posts after a game. But it’s not the mic drop, boom roasted, tool some people think they are. Whether people believe me or not, I don’t care, but I get lucky a few times a year and get to have conversations with players past and present in non official events and to hear them talk about a player compared to how the stats people talk about a player, the difference is freaking wild.

 

[Fitzy]

I think they’re useful as well to an extent. Some help paint a fuller picture, like Valiquettes. I love seeing his posts after a game. But it’s not the mic drop, boom roasted, tool some people think they are. Whether people believe me or not, I don’t care, but I get lucky a few times a year and get to have conversations with players last and present in non official events and to hear them talk about a player compared to how the stats people talk about a player and the difference is freaking wild.

They certainly are not. It's why when someone claims that a player is the "worst in the league" and all they provide is a chart I'm amused.

Even Valiquette, good as he is, sometimes makes assumptions based on data that would lose him points even the simplest quantitative research methods class in an undergrad course. Including ignoring unmeasured qualitative differences between chances that get lumped in the same statistical categories.

 

[Pawnee Rangers]

They certainly are not. It's why when someone claims that a player is the "worst in the league" and all they provide is a chart I'm amused.


View attachment 846612
Even Valiquette, good as he is, sometimes makes assumptions based on data that would lose him points even the simplest quantitative research methods class in an undergrad course. Including ignoring unmeasured qualitative differences between chances that get lumped in the same statistical categories.

Maybe I’m old school or a boomer (I’m not) but I feel like I can get a good grasp on where a team or player is trending just by watching the game. And if Valiiquette’s stuff can be flawed and he actually pays people to watch the games, WTF would I hang in Dom from the Athletics every word?

 

[Fitzy]

Maybe I’m old school or a boomer (I’m not) but I feel like I can get a good grasp on where a team or player is trending just by watching the game. And if Valiiquette’s stuff can be flawed and he actually pays people to watch the games, WTF would I hang in Dom from the Athletics every word?

This, to be honest, is my biggest gripe with stats right now.

The larger the sample size, the more can be gleaned from data, and the more generalizable and reliable it becomes. Small sample sizes like a particular game are open to major issues like random error, and also any small error with how their performance is tracked is magnified over a small game.

So for me, the individual player cards for an individual game? At worst they're useless. At best, they usually tell us exactly what our eye test did about the performances.

It's over like a 50 game sample that I start taking them more seriously.

 

[redwhiteandblue]

Funny hearing Brian Boyle talk about how highly regarded Goodrow is and how bubble teams around the league would love to have a guy who can play within a defensive structure like him. He must be an idiot too.

Bodes well for us if we want to trade him though

Been my point all along. I know since 2 years ago RangerBoy has been beating the drum that he will be bought out during the offseason, but I just don't buy it. There will be a team out there that will take on a SC Champ, and doesn't have a great read on him.

Hell they probably won't until they've had him for 5 or 6 games haha.

 

[Pawnee Rangers]

And @Fitzy to answer your original question, I’m giving the codes to the dude who’s been in live combat and has seen the horrors of war first hand

 

[SA16]

They certainly are not. It's why when someone claims that a player is the "worst in the league" and all they provide is a chart I'm amused.


View attachment 846612
Even Valiquette, good as he is, sometimes makes assumptions based on data that would lose him points even the simplest quantitative research methods class in an undergrad course. Including ignoring unmeasured qualitative differences between chances that get lumped in the same statistical categories.

That is not surprising considering his background is "NHL Goalie" and not stats/math major.

 

[HatTrick Swayze]

True. In theory this isn't between any rando here and Boyle, though. It's between a few of the people doing the statistical work behind the scenes vs Boyle.

So it's more like the question- who would you rather have access to the nuclear launch codes: an eccentric rational choice theorist, or a former general?

I used to be pretty anti metric in the early days, 10 years or so ago. Back in the early days of Corsi when there was zero consideration for shot quality, for example, there were logic gaps big enough to drive a 747 through.

That said I have definitely come around over the years and consider them a valuable resource. By keeping an open mind I have found that they have challenged many of my preconceptions held and developed through my amateur playing career (not at a very high level of course)…basic and IMO non controversial things like spending more time in your opponents zone is a great way to defend.

Now when I watch games I often look at the metrics in real time and ask myself if they confirm or counter what I thought I just observed. Vastly more often than not they align to my perception of what I just watched. I would challenge any skeptics to test out the same exercise.

I have absolutely no doubt that Boyle has a TON of experience and insight to offer. Insight I could never dream of possessing as someone who never played at the D1 level let alone the NHL. But I do wonder how much, if any, time he has spend challenging his preconceived assumptions instilled in him 10-30 years ago. Maybe he has - I have no idea. But as we know this sport has changed a lot in a short period of time. And so often the support in the media or from team sources for players that don’t fit the current analytics narrative seems knee jerk and often tribal. Which is what I often find grating and insufficient.

Now that said I am purely talking about basic “counting” statistics like #of shot attempts, scoring chances, and expected goals. When we get into a predictive model made by some guy on Twitter I take it with a heft grain of salt.

 

[Hunter Gathers]

Lmfao. Devils fans are easily the most insecure fans on the planet. They are more concerned about laughing emojis than their team being dogshit.

Funny hearing Brian Boyle talk about how highly regarded Goodrow is and how bubble teams around the league would love to have a guy who can play within a defensive structure like him. He must be an idiot too.

Bodes well for us if we want to trade him though

Boyle must be a dummy.

That is not surprising considering his background is "NHL Goalie" and not stats/math major.

Just the nature of becoming too reliant on stats, AI, etc.

His model is weird. That model is not very good. This one looks bad because the team was defending a lead late and this model hates that. Don’s model does this. Her model does that. It’s just people looking for relevance in a crowded field of analysts. Seems pretty fugazi.

There is so much f***ing noise in the space right now it's not even funny. So much of this is just "garbage in, garbage out."

That's the problem with any type of modeling for anything. Modeling is only as good as the source material is. So if the source material isn't really all that relevant to what the model is trying to present, well . . .

 

[Hunter Gathers]

I used to be pretty anti metric in the early days, 10 years or so ago. Back in the early days of Corsi when there was zero consideration for shot quality, for example, there were logic gaps big enough to drive a 747 through.

That said I have definitely come around over the years and consider them a valuable resource. By keeping an open mind I have found that they have challenged many of my preconceptions held and developed through my amateur playing career (not at a very high level of course)…basic and IMO non controversial things like spending more time in your opponents zone is a great way to defend.

Now when I watch games I often look at the metrics in real time and ask myself if they confirm or counter what I thought I just observed. Vastly more often than not they align to my perception of what I just watched. I would challenge any skeptics to test out the same exercise.

I have absolutely no doubt that Boyle has a TON of experience and insight to offer. Insight I could never dream of possessing as someone who never played at the D1 level let alone the NHL. But I do wonder how much, if any, time he has spend challenging his preconceived assumptions instilled in him 10-30 years ago. Maybe he has - I have no idea. But as we know this sport has changed a lot in a short period of time. And so often the support in the media or from team sources for players that don’t fit the current analytics narrative seems knee jerk and often tribal. Which is what I often find grating and insufficient.

Now that said I am purely talking about basic “counting” statistics like #of shot attempts, scoring chances, and expected goals. When we get into a predictive model made by some guy on Twitter I take it with a heft grain of salt.

This is mostly where I am at, too. But the problem is we now have competing analytics. These guys are all trying to make money for themselves. So, let's make sure we keep that in perspective. Hopefully the best wins out but that doesn't always happen when you start factoring money into the situation.