Hockey Analytics

[TrueBlueGED]

I know many (most?) around here simply won't care, but today War on Ice announced it will be closing at the end of March. This sucks. By far the best place to go for stats-based research.

[darksabre]

Man, what's the deal with that?

[TrueBlueGED]

Man, what's the deal with that?

Certainly sounds like at least one of the people running it got hired by a team.

[darksabre]

Certainly sounds like at least one of the people running it got hired by a team.

[WildCard]

I know many (most?) around here simply won't care, but today War on Ice announced it will be closing at the end of March. This sucks. By far the best place to go for stats-based research.

This is just depressing  :(

[Brawndo]

I dunno, Stan Bowman has been pretty open that they have proprietary stuff that not only won't he give the specifics on, but he won't even disclose generally what they're about. That tells me the use is pretty significant.

Bowman brought his analytics guy to the first meeting for Team North America, needless to say Chiarelli and company were impressed.

Edited January 25, 2016 by BRAWNDO

[TrueBlueGED]

Nicholas W. Goss ‏@NickGossNESN  5h5 hours ago

Bill James on people who resist analytics in 2016: You used to have to pay attention to those guys. Now you don't. #Boom #SSAC16

 

Maybe true in baseball...

 

On a more serious note, this is undeniably awesome:

 

Fluto Shinzawa ‏@GlobeFluto  9m9 minutes ago

Ducks are using near-field electromagnetic ranging (NFER) to track player movement and metrics. Proprietary info.

[Stoner]

Certainly sounds like at least one of the people running it got hired by a team.

That's the end game, no?

[Brawndo]

Blogger turned Edmonton Oilers exec Tyler Dellow on stats vs analytics: "I started getting paid for it when I called it analytics." #SSAC16

[WildCard]

That's the dream right there. I hear if you hit 100k posts the Sabres hire you. Time to post all night

Edited March 14, 2016 by WildCard

[IKnowPhysics]

On a more serious note, this is undeniably awesome:

 

Fluto Shinzawa ‏@GlobeFluto  9m9 minutes ago

Ducks are using near-field electromagnetic ranging (NFER) to track player movement and metrics. Proprietary info.

 

Novel application of some old physics combined with some interesting new electronics.  One could electronically record where all of their team's players are at all times on the home ice.  It'll be interesting to see if the Ducks are capable of reducing all of that data.  Not a bad way to start though.

[That Aud Smell]

Novel application of some old physics combined with some interesting new electronics.  One could electronically record where all of their team's players are at all times on the home ice.  It'll be interesting to see if the Ducks are capable of reducing all of that data.  Not a bad way to start though.

 

That is pretty cool.

 

I'm wondering when that same (or a similar) technology will extend to the pucks and/or players' sticks. 

[That Aud Smell]

Saw it pointed out on Twitter that, prior to this season, the Penguins hired the War on Ice guy to head up a newly formed analytics department.

 

Crosby, Letang, Malkin - sure.

 

But as is always the case, it's the supporting cast that puts a team over the edge and deep into the playoffs.

 

I have no idea whether the #fancystats kids are arguably behind Pittsburgh's success this year, though.

 

Are they?

[LGR4GM]

Saw it pointed out on Twitter that, prior to this season, the Penguins hired the War on Ice guy to head up a newly formed analytics department.

 

Crosby, Letang, Malkin - sure.

 

But as is always the case, it's the supporting cast that puts a team over the edge and deep into the playoffs.

 

I have no idea whether the #fancystats kids are arguably behind Pittsburgh's success this year, though.

 

Are they?

Matthew Murray is partly behind it. Fleury is god awful in the playoffs and a main reason Pittsburgh hasn't seen the cup finals in 8 years. 

[That Aud Smell]

Matthew Murray is partly behind it. Fleury is god awful in the playoffs and a main reason Pittsburgh hasn't seen the cup finals in 8 years. 

 

I'm sure that's part of it. But, taking a quick peek, MAF was bad for four consecutive playoffs after they won the Cup, but he was not bad the last two. Yes? I know for sure that last year he played well, but got no goal support.

 

Anyway. I'm curious to know whether there's a cause-effect between the fancystat kid(s) and any of the players now making a difference for the Penguins.

[rakish]

In February, 538 had an article about why the Canadian teams weren't going to make the playoffs.  Their answer was, basically, luck. I've been thinking about writing a reply to that article because the question isn't best framed as Canadian vs American, but as old-school vs. analytical (you wouldn't think a website about data would miss this).  Canada teams generally have more old-school vision, and it's killing them.  It's not just Canada though, New Jersey is old-school, Colorado is old-school.

 

But your question, smell, (I use lower case since you made fun of my food porn, now the phrase 'Judge smell cleared the courtroom' makes more sense), is more difficult to answer because what Pittsburgh is using is black box. They're not going to tell you what they know, and how they know it. Why I like working in the draft is that I can tell who knows what, and Pittsburgh generally sucks at it. Maybe with their new people they will be good. Toronto is suddenly very good, New Jersey is very bad.

 

That doesn't mean 'analytics are good,' good analytics are good. One of the teams on the forefront of analytics was the Buffalo Sabres. Regier started trading for high Corsi players, Ruff got Stafford to shoot more, on the belief that since all shooting percentage was luck, it didn't matter where you shot from, it only mattered that you took more shots. This thinking is, and was, ridiculous, and it got Regier and Ruff eventually fired, because Stafford can't score from where Ruff was getting him to shoot from.  Hockey is analytical but it's also knuckle-dragging, Regier put together a team with Luke Adam as the muscle, and that gets your goaltender run.  Regier also used analytics in drafting.  They decided at some point that, Czech's, I think it was, were better draft picks.  So each draft Regier would pick a Czech in the 6th round or so. While there is a tiny bit of truth to the assertion that Czech's are good draft picks, it doesn't mean you can indiscriminately pick Czechs, again it was Regier using analytics terribly.

[That Aud Smell]

difficult to answer because what Pittsburgh is using is black box. They're not going to tell you what they know, and how they know it. 

 

Isn't that every team that's using analytics at this point? Chicago comes to mind.

 

 

Why I like working in the draft is that I can tell who knows what, and Pittsburgh generally sucks at it. Maybe with their new people they will be good. Toronto is suddenly very good, New Jersey is very bad.

 

 

The who now?

 

That doesn't mean 'analytics are good,' good analytics are good. One of the teams on the forefront of analytics was the Buffalo Sabres. Regier started trading for high Corsi players, Ruff got Stafford to shoot more, on the belief that since all shooting percentage was luck, it didn't matter where you shot from, it only mattered that you took more shots. This thinking is, and was, ridiculous, and it got Regier and Ruff eventually fired, because Stafford can't score from where Ruff was getting him to shoot from.  Hockey is analytical but it's also knuckle-dragging, Regier put together a team with Luke Adam as the muscle, and that gets your goaltender run.  Regier also used analytics in drafting.  They decided at some point that, Czech's, I think it was, were better draft picks.  So each draft Regier would pick a Czech in the 6th round or so. While there is a tiny bit of truth to the assertion that Czech's are good draft picks, it doesn't mean you can indiscriminately pick Czechs, again it was Regier using analytics terribly.

 

I don't often equate "early mis-users" with being on the "forefront" of something.

 

But your question, smell, (I use lower case since you made fun of my food porn, now the phrase 'Judge smell cleared the courtroom' makes more sense)

 

:lol:

 

Sorry to have made any fun. The picture of the Good Morning Burger today was enticing.

Edited June 1, 2016 by That Aud Smell

[rakish]

"every team uses it": analytics is a broad term. Both me and Blue have stated that if we ran an organization, we would hire a bunch of people to watch hockey to obtain data. He thinks there's something to be gained by scraping the reports that the NHL produces, I generally don't.  What I'm trying to say is that there's a wide spectrum of 'analytics,' I think Anaheim and Chicago do more, Pittsburgh and Minnesota do some, Calgary and Colorado do none.  I might be totally wrong about Pittsburgh though, they may be doing really good work that's showing up on the ice, hard to tell.

 

working was a poorly chosen word, studying

 

yes, early mis-users would be the better word than forefront.

[WildCard]

Saw it pointed out on Twitter that, prior to this season, the Penguins hired the War on Ice guy to head up a newly formed analytics department.

 

Crosby, Letang, Malkin - sure.

 

But as is always the case, it's the supporting cast that puts a team over the edge and deep into the playoffs.

 

I have no idea whether the #fancystats kids are arguably behind Pittsburgh's success this year, though.

 

Are they?

The Wild hired the WAR on ice founders, not the Penguins. 

In February, 538 had an article about why the Canadian teams weren't going to make the playoffs.  Their answer was, basically, luck. I've been thinking about writing a reply to that article because the question isn't best framed as Canadian vs American, but as old-school vs. analytical (you wouldn't think a website about data would miss this).  Canada teams generally have more old-school vision, and it's killing them.  It's not just Canada though, New Jersey is old-school, Colorado is old-school.

 

But your question, smell, (I use lower case since you made fun of my food porn, now the phrase 'Judge smell cleared the courtroom' makes more sense), is more difficult to answer because what Pittsburgh is using is black box. They're not going to tell you what they know, and how they know it. Why I like working in the draft is that I can tell who knows what, and Pittsburgh generally sucks at it. Maybe with their new people they will be good. Toronto is suddenly very good, New Jersey is very bad.

 

That doesn't mean 'analytics are good,' good analytics are good. One of the teams on the forefront of analytics was the Buffalo Sabres. Regier started trading for high Corsi players, Ruff got Stafford to shoot more, on the belief that since all shooting percentage was luck, it didn't matter where you shot from, it only mattered that you took more shots. This thinking is, and was, ridiculous, and it got Regier and Ruff eventually fired, because Stafford can't score from where Ruff was getting him to shoot from.  Hockey is analytical but it's also knuckle-dragging, Regier put together a team with Luke Adam as the muscle, and that gets your goaltender run.  Regier also used analytics in drafting.  They decided at some point that, Czech's, I think it was, were better draft picks.  So each draft Regier would pick a Czech in the 6th round or so. While there is a tiny bit of truth to the assertion that Czech's are good draft picks, it doesn't mean you can indiscriminately pick Czechs, again it was Regier using analytics terribly.

I could hug you

 

"every team uses it": analytics is a broad term. Both me and Blue have stated that if we ran an organization, we would hire a bunch of people to watch hockey to obtain data. He thinks there's something to be gained by scraping the reports that the NHL produces, I generally don't.  What I'm trying to say is that there's a wide spectrum of 'analytics,' I think Anaheim and Chicago do more, Pittsburgh and Minnesota do some, Calgary and Colorado do none.  I might be totally wrong about Pittsburgh though, they may be doing really good work that's showing up on the ice, hard to tell.

 

working was a poorly chosen word, studying

 

yes, early mis-users would be the better word than forefront.

The Wild you fancy stats a ton. They have been for awhile 

[That Aud Smell]

The Wild hired the WAR on ice founders, not the Penguins. 

 

Looks like the Wild did so after the Penguins did it first:

 

http://nhl.nbcsports.com/2015/07/22/fancy-that-pens-hire-war-on-ices-sam-ventura/

[WildCard]

Looks like the Wild did so after the Penguins did it first:

 

http://nhl.nbcsports.com/2015/07/22/fancy-that-pens-hire-war-on-ices-sam-ventura/

Ah so the Wild hired the other founders

 

http://minnesota.cbslocal.com/2016/01/26/minnesota-wild-hire-stats-specialists-for-hockey-operations/

[Claude_Verret]

Anyone read this one?  Looks interesting.

 

 

Stat Shot 

 

 

[That Aud Smell]

Not yet, but now this will be stuck in my head all day.

 

[That Aud Smell]

In looking at that Travis Yost's (sp?) Twitter feed, I saw him jawing about "micro stats."

 

Anyone know anything about those?

 

Sounds like even some of fancy stat types hate micro stats.

[North Buffalo]

Interesting article on Florida Panthers' new coach and analytics:  

 

https://www.nhl.com/news/tom-rowe-of-florida-panthers-still-learning-evolving/c-284463382