25 under 25 - HF Edition - #4

[JojoTheWhale]

You're talking about great defense masking the performance of a mediocre goalie or something like that, which gives the illusion of having a better goalie than you do. In reality the two things (defense, goaltending) are intertwined but it doesn't change the value of the goalie. Like if Carey Price is 12 goals a year better than an average goalie, he's +12 regardless of his defense. If his defense is +20 goals better than the next team, the team allows 32 fewer goals, but that doesn't change Carey Price's value.

If you're determining Goalie value in a frame of +/- Goals vs average then Shot Quantity/Quality in front of him absolutely affects that. We know that Goalies that face more Shots have a higher SVP. We know that Goalie play is more effectively predicted by HDSVP vs M/LDSVP. We can account for some of these things, but that's not a measurement that lends itself to a +/- metric with the current data that we have available.

 

[FLYguy3911]

1. Sanheim isn't going to flop, he's already a successful NHL player.
2. Even if he did flop, it doesn't really impact Carter Hart.

You're talking about great defense masking the performance of a mediocre goalie or something like that, which gives the illusion of having a better goalie than you do. In reality the two things (defense, goaltending) are intertwined but it doesn't change the value of the goalie. Like if Carey Price is 12 goals a year better than an average goalie, he's +12 regardless of his defense. If his defense is +20 goals better than the next team, the team allows 32 fewer goals, but that doesn't change Carey Price's value.

1. If you're eliminating the possibility that he can or will flop, how is he not the runaway winner of this poll? I'm under the assumption that people are weighing the risk that he may not reach his potential.

2. How does one of our most talented defenseman not panning out NOT impact Carter Hart? The future of the blueline is very much resting on the shoulders of Provorov, Ghost, Sanheim, and Myers to a lesser extent.

Jojo touched on this but shot attempts are not created equally and shots attempts are more frequent with certain players on the ice. Goalie performance is very much influenced by the guys playing in front of them. It's very difficult to truly measure a goalie's worth with expected +/-. And not to speak for everyone, but I doubt that is what people's thought processes were when they voted for Hart. I think most people expect this team to be good and they expect Hart to be THE GUY in net. And they possibly overrated the goaltending position.

I was hoping to get a clearer answer on what the obvious difference between Sanheim and Hart were from a prospect standpoint.

 

[baudib1]

If you're determining Goalie value in a frame of +/- Goals vs average then Shot Quantity/Quality in front of him absolutely affects that. We know that Goalies that face more Shots have a higher SVP. We know that Goalie play is more effectively predicted by HDSVP vs M/LDSVP. We can account for some of these things, but that's not a measurement that lends itself to a +/- metric with the current data that we have available.

That's still going to lead the an average goalie allowing a certain number of goals and a greater goalie allowing fewer.

If you're trying to say that better defense will lead Brian Elliott to save X amount of goals but Superstar Goalie A to allow X+ amount of goals, well no, I don't think that's been established (could be wrong).

 

[baudib1]

1. If you're eliminating the possibility that he can or will flop, how is he not the runaway winner of this poll? I'm under the assumption that people are weighing the risk that he may not reach his potential.

2. How does one of our most talented defenseman not panning out NOT impact Carter Hart? The future of the blueline is very much resting on the shoulders of Provorov, Ghost, Sanheim, and Myers to a lesser extent.

Jojo touched on this but shot attempts are not created equally and shots attempts are more frequent with certain players on the ice. Goalie performance is very much influenced by the guys playing in front of them. It's very difficult to truly measure a goalie's worth with expected +/-. And not to speak for everyone, but I doubt that is what people's thought processes were when they voted for Hart. I think most people expect this team to be good and they expect Hart to be THE GUY in net. And they possibly overrated the goaltending position.

I don't know, sue me for liking Hart. I love Sanheim but I would absolutely trade him in a package for Karlsson. I wouldn't trade Hart in anything, certainly not in a package involving a Quick or Price or something.

 

[Magua]

Talking about goals saved when Sanheim was -6. Shaking my head.

 

[FLYguy3911]

If you're trying to say that better defense will lead Brian Elliott to save X amount of goals but Superstar Goalie A to allow X+ amount of goals, well no, I don't think that's been established (could be wrong).

These things are very difficult to measure but Brian Elliott led the league in GAA and SV% playing in a Hitchcock system and behind deep St. Louis teams. He hasn't performed nearly as well away from Hitchcock and the Blues.

I don't know, sue me for liking Hart. I love Sanheim but I would absolutely trade him in a package for Karlsson. I wouldn't trade Hart in anything, certainly not in a package involving a Quick or Price or something.

I like Hart. Everyone likes Hart. I was just hoping to get a clearer understanding of people's philosophies here. I guess it didn't happen.

As an aside, I may be completely wrong, but I don't think Hart has much standalone trade value. Team's would absolutely ask for him in a package, but I don't think he's bringing back much value by himself (risk of 20 year old goalies and such). Sanheim could bring back a decent NHL player, even if the Flyers have done a good job of tanking his trade value.

 

[Striiker]

Talking about goals saved when Sanheim was -6. Shaking my head.

Still blows my mind that people try to cite +/- in 2018.

 

[JojoTheWhale]

That's still going to lead the an average goalie allowing a certain number of goals and a greater goalie allowing fewer.

If you're trying to say that better defense will lead Brian Elliott to save X amount of goals but Superstar Goalie A to allow X+ amount of goals, well no, I don't think that's been established (could be wrong).

Yes, of course a better Goalie would save more over the long run if play in front of the two were magically equal, but it’s defining better (and its degrees) and isolating Goalie play that’s tricky. Allowing less Shots to hit the net isn’t a Goalie skill and our data is woeful.

GoalsSavedAboveAverage (GSAA) is exactly the kind of cumulative +/- system you mentioned. People like Nick Mercadante have taken it an extra step and tried to account for things like Shot Location and turned it into a Rate to minimize the playing time distortion. He called that version AdjustedGSAA/60 and is often quoted specifically for 5v5 situations, but even he openly admits it’s just a baseline for future developments. Not surprisingly, Emmanuel Perry had his own take on GSAA. The write up is on Corsica.

Basically, Goalies are voodoo and so are their stats. Also, if you think a statistical +/- system like this has real value, you should have voted for Steve Mason for the Vezina in 2015-16.

 

[Striiker]

These things are very difficult to measure but Brian Elliott led the league in GAA and SV% playing in a Hitchcock system and behind deep St. Louis teams. He hasn't performed nearly as well away from Hitchcock and the Blues.

I vaguely remember another goalie who put up good numbers when with a different system/team and then looked like trash when he was here.

I think he might have been Russian or something...

 

[Magua]

I vaguely remember another goalie who put up good numbers when with a different system/team and then looked like trash when he was here.

I think he might have been Russian or something...

Brian Elliott is definitely not Russian brah.

 

[Striiker]

Brian Elliott is definitely not Russian brah.

That’s why I said ANOTHER goalie.

I obviously meant Rob Zepp.

 

[baudib1]

Yes, of course a better Goalie would save more over the long run if play in front of the two were magically equal, but it’s defining better (and its degrees) and isolating Goalie play that’s tricky. Allowing less Shots to hit the net isn’t a Goalie skill and our data is woeful.

GoalsSavedAboveAverage (GSAA) is exactly the kind of cumulative +/- system you mentioned. People like Nick Mercadante have taken it an extra step and tried to account for things like Shot Location and turned it into a Rate to minimize the playing time distortion. He called that version AdjustedGSAA/60 and is often quoted specifically for 5v5 situations, but even he openly admits it’s just a baseline for future developments. Not surprisingly, Emmanuel Perry had his own take on GSAA. The write up is on Corsica.

Basically, Goalies are voodoo and so are their stats. Also, if you think a statistical +/- system like this has real value, you should have voted for Steve Mason for the Vezina in 2015-16.

I'm not extremely well-versed in advanced hockey metrics but that's not my point; I'm not arguing about the impact of defensemen on GSAA. I'm positing that the actual value in terms of goals for a great goalie over an average (or below average) goalie doesn't change because they have different defensemen in front of them. Carey Price is worth X amount of goals over Brian Elliott. Obviously their numbers will look different if you have them switch teams.

Anywho; we're probably spoiled by the success of Ghost and Provorov. At any other point in Flyers history we would be so goddamned excited about a homegrown defenseman as good as Sanheim it might be a different story.