A SERIOUS DISCUSSION - The Goalies

[What the Faulk]

It's certainly debatable. I wouldn't call what he has insane, though. À case could be made either way depending on how much you value games played. That's why it's hard to place Khudobin at the moment.

 

[Navin R Slavin]

But I don't go by stats with goalies, I try to watch them as much as I can.

Yup. Me too.

Yesterday I watched a goalie whose glove hand appeared to be encased in cement.

--hank

 

[RodTheBawd]

http://www.quanthockey.com/nhl/seasons/2010s-nhl-goalies-stats.html

Basically if you're using anything other than SV% to judge goalies, you're doing it wrong. Wins and GAA are team stats. Goalies own their SV%'s and the teams they play for have negligible effect on them over the long run. This has been shown and proven numerous times now.

Until a meaningful way to quantify quality of shot comes out.

 

[What the Faulk]

Until a meaningful way to quantify quality of shot comes out.

In the grand scheme of things, shot quality is meaningless.

 

[Ole Gil]

There was some analysis one of those people that does stuff like this did that showed chances off the rush had a higher shooting %. As well as chances on the PP. As well as shots from Forwards rather than defensemen.

It's pretty easy how to see how a teams style and quality of play would impact those, and consequently, the Save%.

Save% is a lazy stat.

(one of the links actually quantifies shot quality's effect as 20-25% of the effect of shot quantity. So it's far from negligible, even in the general sense.)

 

[What the Faulk]

Are you talking about Chris Boyle's work? I definitely thought that was interesting, but there seems to be more evidence in the opposite direction. I'd like to see more people look at his research on a larger sample size instead of just a handful of goaltenders.

 

[BrindAmours Army]

In the grand scheme of things, shot quality is meaningless.

Eh. Your own sources disagree with you.

"I try to clarify the distinction between "doesn't exist" and "isn't as important as people imagine"

"Shot quality isn't insignificant, but it's clearly subordinate."

"You add all these factoids together and it becomes apparent that -while shot quality differences exists- it does not add overwhelming increases in discovery to shot metric analysis."

"In general, shot quality factors tend to be small enough that they don’t grossly alter our understanding of the game"

"The argument is actually that most shot quality effects are smaller than people think, and that over the sample sizes we normally work with, differences in shot quality tend to be dominated more by noise than talent. "

While current shot quality statistics are not nearly as significant as shot differential, they are not meaningless.

Also, the post you responded to:

Until a meaningful way to quantify quality of shot comes out.

Relies on the notion that current factors do not quantify quality of shot in any meaningful way. As such, even had all of the articles you posted stated that quality of shot was insignificant, they would be relying upon current methodology, which would be irrelevant to the quoted post.

In conclusion, I really need to stop procrastinating by arguing semantics and start studying

 

[What the Faulk]

Oh semantics you say? I did preface my admittedly hyperbolic conclusion with "in the grand scheme of things". Checkmate, observists.

 

[RodTheBawd]

For the most part, I tend to choose my words/phrasing pretty deliberately.

 

[cptjeff]

Yup. Me too.

Yesterday I watched a goalie whose glove hand appeared to be encased in cement.

--hank

You also watched (if that is indeed the right word, were you at the game?) a goalie on day three of training camp. It's not promising, but it's hardly much to go on.

 

[Ole Gil]

For the most part, I tend to choose my words/phrasing pretty deliberately.

That's a silly way of communicating.

 

[garret9]

Shot quality is real... but you'll be right more often ignoring it than you will be by adding it due to its uncontrollable nature.
That's the issue with shot quality. Not that it's hard to measure, but that we don't find people ever being able to sustainably control it.
For the most part, large sample save% tends to outperform the eye-test. The issue then is that you need a large sample, which means you can only really compare guys that have survived the eye-test first.

Thanks,
One of the authors of above articles.

 

[Navin R Slavin]

You also watched (if that is indeed the right word, were you at the game?) a goalie on day three of training camp. It's not promising, but it's hardly much to go on.

He could get better. That would be nice.

I could also get younger. That would also be nice.

--hank

(p.s. yes, of course I was at the game. I tend to choose my words/phrasing pretty deliberately.)

 

[tarheelhockey]

For the most part, I tend to choose my words/phrasing pretty deliberately.

I'm not buying this statement until I see some reliable data on your quality of word choice.

 

[RodTheBawd]

That's a silly way of communicating.

One of those habits I can't seem to shake.

I'm not buying this statement until I see some reliable data on your quality of word choice.

I'm sorry, there is currently no data to support word choice quality, therefore, it is nearly entirely useless.

 

[Anton Babchuk]

cam "i'm not good at hockey, in fact i am quite mediocre" ward

 

[Ole Gil]

Are you talking about Chris Boyle's work? I definitely thought that was interesting, but there seems to be more evidence in the opposite direction. I'd like to see more people look at his research on a larger sample size instead of just a handful of goaltenders.

I don't remember who's it was. They were looking at shooting%'s in various situations (rather than goalie sv%). It was something that seemed intuitive put onto paper with numbers assigned. Rushes > Non-Rush. Forward > Defense. PP > EV. And all by several % points. Enough that if you weighted the shots when comparing teams, it'll highlight whether something is an outlier or explainable.

My gut tells me that the actual teams are already doing this, but the people doing it are doing it for a lot of money, and not just handing it out to joe shmoes.

A nice addition to having weighted shots, is that we'd be able to do a better short term analysis. So when a goalie gives up 3 goals on 15 shots, we wouldn't be using expert analysis like "You can't give up 3 goals on 15 shots." It's possible we'd be able to say "They had 2.6 goals worth of shots" That's something that's just not possible in the "Every shot is equal" world.

 

[AD Skinner]

cam "i'm not good at hockey, in fact i am quite mediocre" ward

I'm pretty partial to C4m W4rd- don't know if that one's yours or not but it's a good one

 

[Anton Dubinchuk]

cam "i'm not good at hockey, in fact i am quite mediocre" ward

more like cam "i'm bad at hockey in fact i am quite bad" ward

choose your words more carefully

 

[BrindAmours Army]

Oh semantics you say? I did preface my admittedly hyperbolic conclusion with "in the grand scheme of things". Checkmate, observists.

Well a literalist would treat the idiom in the same way as the hyperbole...Taking the statement as a whole to mean "All things considered shot quality has no significant effect"

Though, fair enough.

Shot quality is real... but you'll be right more often ignoring it than you will be by adding it due to its uncontrollable nature.
That's the issue with shot quality. Not that it's hard to measure, but that we don't find people ever being able to sustainably control it.
For the most part, large sample save% tends to outperform the eye-test. The issue then is that you need a large sample, which means you can only really compare guys that have survived the eye-test first.

Thanks,
One of the authors of above articles.

It isn't hard to measure? Are all relevant factors currently used to determine quality of shot then? I would say we don't have nearly enough data to reliably determine quality of shots. In the perfect world we would have the data for everything relevant which would allow more effective analyses. Shot speed. Angle of trajectory in relation to goaltender's positioning in the net. Presence of an effective screen.

Finding someone that can sustain a high quality of shot isn't really an issue when you're looking to judge goaltender quality. A good shot is a good shot whether it comes from an elite winger or a fourth liner who manages to make all of six good shots per season.

 

[Accelleratii]

I think one goalie will strive as the other falls into mediocrity.

Didn't this already happen?

 

[Unsustainable]

Didn't this already happen?

History has a way of repeating.

 

[Anton Babchuk]

http://www.wralsportsfan.com/ward-seeking-to-regain-no-1-spot/14007277/

mike maniscalco professes his love for cam ward and tells us pleb fans how dumb we are for not understanding ward's greatness. one of the most disgusting puff pieces i've ever seen. *NSFW*

 

[Boom Boom Apathy]

You can’t expect to get starts and get rewarded with ice time because of what you have done in the past. You have to be rewarded for the present and what you bring to the table now. I want to be able to compete and earn that ice time I am going to be given.

Where was that attitude last season?

 

[VAcaniac]

I thought I was a Cam Ward apologist.