Dan Boyle Mod post #696

[DanielBrassard]

Sure. No problem there. Just solely commenting on the use of raw CF% numbers. Of course dCorsi is very dependent on how well they are able to actually estimate what the "expected" performance in a given spot is but it doesn't need to be perfect. As long as it is better than other things that's all that matters. That and that it's constantly being reviewed/updated/tweaked as more info comes out.

Absolutely. Just pointing out for those that don't know that there are some formulas for expected numbers.

 

[Hunter Gathers]

What's the metric that would show how Girardi plays solely against the Triplets versus how others play solely against them?

 

[Raspewtin]

What's the metric that would show how Girardi plays solely against the Triplets versus how others play solely against them?

There's a "matchups" option on war-on-ice that shows a player's metrics vs each individual player if it's specifically them you want to see.

 

[Hunter Gathers]

There's a "matchups" option on war-on-ice that shows a player's metrics vs each individual player if it's specifically them you want to see.

More interested if anyone has compiled this into a graph since I'm pretty interested in seeing how certain players compare when facing a group of guys like that.

 

[Mikos87]

i still think corsi needs to be weighted much better. i think a 46% corsi if youre up against the triplets or something like that isnt nearly as bad as a 46% corsi up against Tanner Glass. they need to weigh the competition in that number...as well as the linemates.

That's looking at it like a rational human being. A good example would be the 4th line guys that CHI uses. Especially in the playoffs.

Some of those guys start 75-80% of their faceoffs in the d-zone.

Another possession king is Bergeron. It helps when the guy is the faceoff king too. Having the puck off of the draw.

And don't forget zone starts don't consider changes on the fly either.

 

[SA16]

Agree with this. Everyone's like "corsi's not perfect so why bother?". It might not be perfect but it's a damn good option. Should everything that's not 100% time tested just be thrown out the window?

Absolutely. Here's a somewhat simple example using baseball as the medium.

Let's say we know that David Wright is starting today and that he will be facing a left handed pitcher who is league average. David Wright has a career 167 wRC+ vs lefties and 122 vs righties. (for the sake of the example let's just pretend this is his true talent vs lefties right now. The numbers themselves really don't matter. It's purely an example).

What will Wright's expected wRC+ be in this particular game? It's clearly not 167. Why? Because he will never see a left handed reliever unless the game is a blowout or it's a closer like Chapman or something. I think from my research in the past on average a starting pitcher faces 24 batters a game and the average game has 38 PA.

So David Wright's expected wRC+ should be something like 167 * (24/38) + 122 * (14/38) = 150. Is this perfect? Of course not. It's probably not a linear effect. The RP he faces on average will be better than SP. There's the TTO penalty. But is it BETTER than what we have otherwise? Undoubtedly yes.

 

[Raspewtin]

More interested if anyone has compiled this into a graph since I'm pretty interested in seeing how certain players compare when facing a group of guys like that.

Blue Blooded a long time ago measured how each of the Rangers' D (when we still had Stralman) performed against different measures of competition (weighted by CF% away from that player) if that answers your question. Otherwise, no someone would have to do all the leg work themself.

 

[SA16]

Blue Blooded a long time ago measured how each of the Rangers' D (when we still had Stralman) performed against different measures of competition (weighted by CF% away from that player) if that answers your question. Otherwise, no someone would have to do all the leg work themself.

The one thing you have to wonder about CF% and other stats is why do they entirely not support what people already think. Like I'm not saying we should try to make stats biased to what we already think but for the most part scouts and experts and just regular people are pretty good at knowing what they are watching.

Ask someone who knows baseball the top 10 baseball hitters and you'll see almost all of them towards in terms of wRC+. But ask someone the top 10 NHL defenseman and you may see none of them near the top in terms of CF% or dCorsi. In fact you may see them towards the bottom. So either everyone who has watched the sport for years and years is just completely wrong and has no idea what they're talking about (I suppose it's possible but you have to admit it's reasonably unlikely) or the stats just aren't quite there yet and need a lot of work.

For example the top 10 defenseman in CF% last year min 615 minutes were Muzzin/McNabb/Rundblad/Doughty/Letang/Stralman/B. Smith/Keith/Leddy/Boychuk.

And then in the top 10 of dCorsi impact last year you have guys like Stralman/Demers/Gardiner/Schlemko (With Ekholm/Michalek/Jackman the next 3) etc... When you see something like that come up it makes it pretty likely that something is not quite right or just this number is really not that useful at all without context.

 

[DanielBrassard]

The one thing you have to wonder about CF% and other stats is why do they entirely not support what people already think. Like I'm not saying we should try to make stats biased to what we already think but for the most part scouts and experts and just regular people are pretty good at knowing what they are watching.

Ask someone who knows baseball the top 10 baseball hitters and you'll see almost all of them towards in terms of wRC+. But ask someone the top 10 NHL defenseman and you may see none of them near the top in terms of CF% or dCorsi. In fact you may see them towards the bottom. So either everyone who has watched the sport for years and years is just completely wrong and has no idea what they're talking about (I suppose it's possible but you have to admit it's reasonably unlikely) or the stats just aren't quite there yet and need a lot of work.

For example the top 10 defenseman in CF% last year min 615 minutes were Muzzin/McNabb/Rundblad/Doughty/Letang/Stralman/B. Smith/Keith/Leddy/Boychuk.

And then in the top 10 of dCorsi impact last year you have guys like Stralman/Demers/Gardiner/Schlemko (With Ekholm/Michalek/Jackman the next 3) etc... When you see something like that come up it makes it pretty likely that something is not quite right or just this number is really not that useful at all without context.

It's a good question, but I wonder if you asked that question 10-15 years ago, if it would sort of be similar. In 2001 I'd guess a guy like Ichiro was considered a top-10 hitter. Is it that unlikely that a certain type of defense was over or undervalued?

 

[SA16]

It's a good question, but I wonder if you asked that question 10-15 years ago, if it would sort of be similar. In 2001 I'd guess a guy like Ichiro was considered a top-10 hitter. Is it that unlikely that a certain type of defense was over or undervalued?

Oh it definitely was. The whole "Jeter" type of defense of a good glove but poor range was seen as like the way to go because the only stats back then were like errors and fielding percentage and that type of player looks great according to those metrics. Who knows though.

 

[silverfish]

i still think corsi needs to be weighted much better. i think a 46% corsi if youre up against the triplets or something like that isnt nearly as bad as a 46% corsi up against Tanner Glass. they need to weigh the competition in that number...as well as the linemates.

This is the first 1600 pages of the two advanced stats threads combined.

 

[bobbop]

This isn't true.

Sorry, but it is true. Bad for most of the regular season, especially bad for 15-20 games after he returned from injury.

But again, not bad at all in the playoffs.

He was better in the playoffs, but he had that terrible game 1 against Washington where he pretty much lost us the game all by himself with very poor defense on multiple goals.

In fairness, the winning goal in that game was scored after he lost the puck when he took an unpenalized head shot.

 

[Mac n Gs]

What's the metric that would show how Girardi plays solely against the Triplets versus how others play solely against them?

You could also use a super WOWY and compare how different players line up against others.

Here's the link:

http://www.puckalytics.com/superwowy.html

 

[RangerBoy]

Does Boyle talk to the print media? It would have been interesting to hear his thoughts on being a healthy scratch and on his play so far. Hockey players are usually really good with the media.

 

[NYRFANMANI]

beat me to it

 

[Waivers]

Three games in and already 'fans' are trying to put Dan Boyle's head on a pike already.

This is the same scenario when Talbot played ~30 games after Hank was hurt, and then was asked, do we actually play one of the world's best goaltenders who make $10,000,000 per year, or have him ride the pine instead.

Let's give him some time and not make irrational decisions so soon.

 

[Hunter Gathers]

Three games in and already 'fans' are trying to put Dan Boyle's head on a pike already.

Boyle wasn't that good last year (decent 4/5). This is a continuation of his aging (now a 5/6).

 

[Amazing Kreiderman]

Boyle wasn't that good last year (decent 4/5). This is a continuation of his aging (now a 5/6).

And yet he did what he was brought here for.

7th overall among D-men in points in the play offs
1st overall among D-men in PP points in the play offs

Reg season PP GF%
Dan Boyle: 5.942 GF/60, 0.743 GA/60
Dan Girardi: 5.627 GF/60, 0 GA/60
Ryan McDonagh: 5.700 GF/60, 0.760 GA/60
Marc Staal: 0 GF/60, 0 GA/60
Kevin Klein: 0 GF/60, 0 GA/60
Keith Yandle: 6.145 GF/60, 0.723 GA/60

Defensively at ES he was not bad either:

5-on-5 stats GA/60
Dan Boyle: 1.255
Dan Girardi: 2.360
Ryan McDonagh: 2.358
Marc Staal: 1.951
Kevin Klein: 2.197
Keith Yandle: 2.776

 

[Captain Lindy]

I have a feeling something will happen within the next few weeks with Boyle. They told Diaz they'd have him back up in a month, or "revisit" his situation or something like that. Sounds like something is cooking.

 

[Filthy Dangles]

Agree with ecl.

I hope this is a precursor to somehow moving Boyle and 4.5 off the books this year. Klein and McIlrath are both better than him at this point.

Rongos gonna Rongo and always acquire players once their time is pretty much up. Would have loved DB in 2008 when we tried to first acquire him but he's pretty much a detriment besides powerplay at this point in his career.

If and when happens, I will be ready for the impending implosion on these boards complaining about the absolute WOAT RHD depth in Girardi, Klein and McIlrath in the history of hockey.

 

[Raspewtin]

And yet he did what he was brought here for.

7th overall among D-men in points in the play offs
1st overall among D-men in PP points in the play offs

Reg season PP GF%
Dan Boyle: 5.942 GF/60, 0.743 GA/60
Dan Girardi: 5.627 GF/60, 0 GA/60
Ryan McDonagh: 5.700 GF/60, 0.760 GA/60
Marc Staal: 0 GF/60, 0 GA/60
Kevin Klein: 0 GF/60, 0 GA/60
Keith Yandle: 6.145 GF/60, 0.723 GA/60

Defensively at ES he was not bad either:

5-on-5 stats GA/60
Dan Boyle: 1.255
Dan Girardi: 2.360
Ryan McDonagh: 2.358
Marc Staal: 1.951
Kevin Klein: 2.197
Keith Yandle: 2.776

You're fighting a losing battle, don't bother.

 

[Amazing Kreiderman]

People will die trying to defend Staal, Girardi, Lundqvist, Stepan and Kreider but when a player is acquired trough free agency or a trade they have to play like Bobby Orr/Wayne Gretzky to avoid being criticized.

 

[Bleed Ranger Blue]

Theres certainly more than meets the eye here. I refuse to believe Boyle is sitting simply because he played a poor 3 games to start the season. Sure sounds like the whispers of friction with the coaching staff has some legs.

The Rangers have some bad luck with 39 year olds that "want to play for the Rangers, and Rangers only." How about, one of these days, the Rangers don't jump at that sort of chance.

 

[Raspewtin]

Theres certainly more than meets the eye here. I refuse to believe Boyle is sitting simply because he played a poor 3 games to start the season. Sure sounds like the whispers of friction with the coaching staff has some legs.

The Rangers have some bad luck with 39 year olds that "want to play for the Rangers, and Rangers only." How about, one of these days, the Rangers don't jump at that sort of chance.

I don't know, he has been really bad these last three games. I don't think it's that ridiculous.

 

[Filthy Dangles]

People will die trying to defend Staal, Girardi, Lundqvist, Stepan and Kreider but when a player is acquired trough free agency or a trade they have to play like Bobby Orr/Wayne Gretzky to avoid being criticized.

Funny. How about Anton Stralman, the guy he essentially replaced on the roster, for starters.

I would love to eat all this crowe and see Dan Boyle contribute with a 40 point season, top 10 NYR PP and respectable defense (easing the dzone start load bore by #5, 27 and #18) in a top4 role en route to a long playoff run that ends in the parade, but I can't see it with Boyle at his point.