GDT 11/26 GDT : Philadelphia Flyers 2 at Red Wings 5

[GoWings1905 2,694]

Jurco was also terrific last night. I thought his best outing of the season and he has five points in five games now.

Admittedly, I didn't originally like the idea of Jurco with Glendening and Miller, but he seemed to feed off their energy.

[rick zombo 3,739]

Jurco was also terrific last night. I thought his best outing of the season and he has five points in five games now.

Admittedly, I didn't originally like the idea of Jurco with Glendening and Miller, but he seemed to feed off their energy.

Jurco was great on that line. He's a big-body who skates waaaay better than Andersson and instantly make that line a threat to score.

[Mckinley25 677]

You realize that advanced statistics are based on what actually happens on the ice right?

Shut up with that intellectual mumbo jumbo gypsy talk!!

[frankgrimes 1,836]

You realize that advanced statistics are based on what actually happens on the ice right?

They are based on past examples, given some funny names and done by math freaks and excel geniuses but I guess I'm just not advanced enough to care about stats other than goals, assists, gave is... the official ones. And even then they aren't telling the whole story a player could have a great game but still be snake bitten

Edited November 27, 2014 by frankgrimes

[e_prime 1,936]

I did not watch that broadcast but maybe Clement should do some research. Or not use information from 3 years ago.

http://stats.nhlnumbers.com/teams

NJD 30.407 PIT 28.461 MTL 28.396 PHI 28.353 NYR 28.132 NAS 27.918 DET 27.896 COL 27.829 VAN 27.817 FLA 27.803

Hilarious. Philly actually has an older team than us. Maybe Clement needs to be reminded he's doing a national broadcast...

[nawein 324]

They are based on past examples, given some funny names and done by math freaks and excel geniuses but I guess I'm just not advanced enough to care about stats other than goals, assists, gave is... the official ones. And even then they aren't telling the whole story a player could have a great game but still be snake bitten

And that, usually, can be shown using advance stats. That's why they're there. To tell more of the story.

[KATIEBARTHEDOOR24 426]

They are based on past examples, given some funny names and done by math freaks and excel geniuses but I guess I'm just not advanced enough to care about stats other than goals, assists, gave is... the official ones. And even then they aren't telling the whole story a player could have a great game but still be snake bitten

"Statistics are for losers, unless they are good ones" Mickey Redman

[Euro_Twins 4,475]

So, like, this stuff is anti-hockey?

Stats don't matter. That's why Cleary is a superstar

[Dabura 12,207]

You guys are kidding, right? By far THE WORST. Clement is still stuck on how old our team is... "Oldest team in the league." Really, Bill. What about NJ where 95% of their forward core is 30+

...and talking about how we over-ripen our prospects. No, duh Bill. This is how it works in Detroit.

I mean, we could be Philly where we just trade everyone away before anyone gets any time at all. BLOCKBUSTER TRADES!!!

As far as the game goes... we really need to figure out the third period. There was a hell of a lot of scrambling. First period, poor. Second period, good. Third period, poor.

I couldn't believe he - whoever it was - actually said, straight up, Mike Babcock will definitely not be re-signing with the Wings because they're old and their future is really not looking bright right now. And on a national telecast! And about that - if I'm watching an NBC telecast, I'm kind of expecting, y'know, an NBC telecast.

As for our third periods - I chalk it up to "score effects" more than anything else. A team with a decent lead is usually going to ease its foot off the gas and the team that's trailing is going to pick up its game (which you'll likely see reflected in the shot count). If you go into the third period with a three-goal separation, you're almost definitely going to see it happen. The trailing team thinks, "Hey, we've got nothing to lose. Let's just throw everything we've got at them." Look at Ottawa a couple nights ago - they damn near sent the game into OT.

For what it's worth, though, we're +5 in first period (15 goals for, 10 goals against), +7 in the second (25/18), and +1 in the third (22/21).

[haroldsnepsts 4,826]

Weiss was our best forward tonight in terms of Corsi-for percentage. That, to me, is even more encouraging than the three goals in two games.

I'm not against advanced stats, but using Corsi for individual players is pretty shaky to me. All this comes with the caveat that I'm not super versed in advanced stats, but from what I know Corsi is already built on a somewhat creaky premise, but one that usually works out for a team over a significant number of games. Comparing individual players though I think it starts to break down. It's really a dressed up plus minus stat.

If I understand it correctly (and it's quite possibly I don't), Weiss having the best Corsi-for percentage last night means when he was on the ice at even strength, the team got off more Shots For than they allowed against. As for what that says about his play last night? To me not a whole lot, especially when you factor in sample size.

by any metric though it sounds like he's doing well, which is great.

[Hockeytown0001 7,652]

Very glad to see Weiss back on track.

[Dabura 12,207]

They are based on past examples, given some funny names and done by math freaks and excel geniuses but I guess I'm just not advanced enough to care about stats other than goals, assists, gave is... the official ones. And even then they aren't telling the whole story a player could have a great game but still be snake bitten

http://www.nhl.com/ice/news.htm?id=727093

Detroit Red Wings coach Mike Babcock is of the belief that the more information he can gather the better he will be at evaluating and running his team. That's why Babcock is interested in the growing movement of advanced statistical analysis in the NHL.

"Not only is it a great idea, but if you don't [start using analytics] you're going to fall behind," Babcock told NHL.com. "You have to be on the cutting edge. It was [Arizona Coyotes assistant general manager] Darcy Regier who said, 'If you didn't invent it, you have to be the second- or third-best copier, because if you're fourth or fifth you've got no chance.'"

Babcock said the Red Wings have not hired someone to serve as an analytics guru, but he said such a move is coming.

"Not quite yet, but 'yet' is the key word there," he said. "We will, for sure. We just have to."

The Boston Bruins, Toronto Maple Leafs and New Jersey Devils this summer have hired or promoted from within executives with statistical backgrounds. Pittsburgh Penguins general manager Jim Rutherford has talked about how he will incorporate analytics into the front office.

The Chicago Blackhawks and Los Angeles Kings have used analytics and have won the Stanley Cup twice each in the past five seasons. St. Louis Blues coach Ken Hitchcock is on record saying he uses analytics as part of his decision-making process for matchups and chemistry.

"I've seen so many analytics presentations it's not even funny," Babcock said. "Some of them have been very impressive and some of them I wasn't as impressed with, but I love the information. I absolutely love the information. We're in the information business, so how do you get it? And then you have to use your expertise to sort it out."

A basic element of the analytics movement happening in the NHL is possession-based stats featuring shot attempts-for versus shot attempts-against in 5-on-5 situations. Understanding the positive effects of possession and shooting the puck is not new to the hockey community, but stats found through possession metrics Corsi and Fenwick are relatively new and are being examined for their usefulness and effectiveness.

The game is all about possession. That much isn't really up for debate. The problem is, we don't have a way of measuring time of possession in hockey. So we use a proxy: shot attempts. The idea being, you can only shoot the puck if you possess the puck. Corsi is basically just +/- for shot attempts. Fenwick is the same deal except you exclude blocked shots. Sounds simple, and it is, but there's great value here. Advanced stats (which should just be called something less off-putting. Maybe extended stats?) aren't voodoo math. It's about getting a better (i.e. more nuanced, more quantifiable, more precise, more exact, more to-the-heart-of-what's-really-going-on in a given situation) handle on the game and turning that into a competitive advantage.

Edited November 27, 2014 by Dabura

[Dabura 12,207]

I'm not against advanced stats, but using Corsi for individual players is pretty shaky to me. All this comes with the caveat that I'm not super versed in advanced stats, but from what I know Corsi is already built on a somewhat creaky premise, but one that usually works out for a team over a significant number of games. Comparing individual players though I think it starts to break down. It's really a dressed up plus minus stat.

If I understand it correctly (and it's quite possibly I don't), Weiss having the best Corsi-for percentage last night means when he was on the ice at even strength, the team got off more Shots For than they allowed against. As for what that says about his play last night? To me not a whole lot, especially when you factor in sample size.

by any metric though it sounds like he's doing well, which is great.

It doesn't tell us much, but it's encouraging in that his possession numbers have been really, really lousy in his short time as a Wing. If he can consistently put up good possession numbers, that'll be great. And with an appropriately large sample size, we could look at him through the lens of a With Or Without You (WOWY) analysis.

I'm actually not 100% sold on the almighty power of advanced stats. They definitely have their limitations, and some of their biggest proponents are in fact evil math freak pricks who think these metrics are the be-all and end-all of hockey analysis. But I do believe in them (the metrics, not the evil math freaks), and I think there's real value in them.

[Dabura 12,207]

http://www.hockeywilderness.com/2014/9/25/6843523/wilderness-walk-zach-parise-stat-nerd-edition

[Son of a Wing 1,644]

http://www.hockeywilderness.com/2014/9/25/6843523/wilderness-walk-zach-parise-stat-nerd-edition

Did frankgrimes write that?

[haroldsnepsts 4,826]

It doesn't tell us much, but it's encouraging in that his possession numbers have been really, really lousy in his short time as a Wing. If he can consistently put up good possession numbers, that'll be great. And with an appropriately large sample size, we could look at him through the lens of a With Or Without You (WOWY) analysis.

I'm actually not 100% sold on the almighty power of advanced stats. They definitely have their limitations, and some of their biggest proponents are in fact evil math freak pricks who think these metrics are the be-all and end-all of hockey analysis. But I do believe in them (the metrics, not the evil math freaks), and I think there's real value in them.

That's the problem I have with using Corsi for an individual player though, especially in one game. Corsi doesn't track possession at all. It tracks shots that occur when a player is on the ice at even strength. For a team over a large number of shots, it appears to correlate pretty well with possession time. I don't know how well that correlates with possession time for one player in one game.

For a single player over the course of a game, it's too plus/minusy for me. When Weiss was out there even strength, there were more shots on the other teams net. But then there's quality of linemates, quality of opponents, offensive vs. defensive zone starts. WOWY analysis helps a bit but at a certain point you have to corroborate with so many other stats that is it really telling us that much?

I agree overall that the "advanced" stats can be useful. But they're already being misused all the time with people making claims that aren't actually supported by the stats. It's a bit of a new toy.

[Echolalia 2,961]

That's the problem I have with using Corsi for an individual player though, especially in one game. Corsi doesn't track possession at all. It tracks shots that occur when a player is on the ice at even strength. For a team over a large number of shots, it appears to correlate pretty well with possession time. I don't know how well that correlates with possession time for one player in one game.

For a single player over the course of a game, it's too plus/minusy for me. When Weiss was out there even strength, there were more shots on the other teams net. But then there's quality of linemates, quality of opponents, offensive vs. defensive zone starts. WOWY analysis helps a bit but at a certain point you have to corroborate with so many other stats that is it really telling us that much?

I agree overall that the "advanced" stats can be useful. But they're already being misused all the time with people making claims that aren't actually supported by the stats. It's a bit of a new toy.

I agree that the next generation of statistics is going to be a great opportunity for teams to improve and maximize their chances at scoring, but at the same time you bring up a good point about the the situations surrounding the stats. Of course player X will have a good corsi-for, because coach only plays him for offensive zone faceoffs.

Ultimately, these statistics are new tools in our toolbag, but we have to know how and when to use them. Its like having a fancy new power drill when you used to have only screwdrivers and hammers. But you're still not going to use the drill if you need to something sawed. That's where I see the biggest issue going forward, and moreso with fans than the guys that are being hired to analyze these stats (hopefully with advanced statistical education). We've already seen lots of people try sawing with a drill. And I doubt it'll get much better in the recent future.

And then there's the other types who reject the fancy stats alltogether. The power drill uses electricity, and comes with a convoluted instruction manual (only in Mexican). Its confusing and seems dangerous. They'd rather stick with the good ol' fashioned hammer and screwdriver.

[kliq 3,755]

They are based on past examples, given some funny names and done by math freaks and excel geniuses but I guess I'm just not advanced enough to care about stats other than goals, assists, gave is... the official ones. And even then they aren't telling the whole story a player could have a great game but still be snake bitten

Aren't all stats based on on past examples? If you are not evaluating a player based on past examples, you are evaluating based on future projections. However future projections are based on past examples.

Advanced stats give GM's and coaches more tools to work with when evaluating talent. I am not saying that these stats are a greater tool then watching a player play, but Holland, Babcock and company cant watch every player in the league for the entire game, every game of the year. When making a trade and/or signing a free agent, these stats should be considered.

[frankgrimes 1,836]

Aren't all stats based on on past examples? If you are not evaluating a player based on past examples, you are evaluating based on future projections. However future projections are based on past examples.

Advanced stats give GM's and coaches more tools to work with when evaluating talent. I am not saying that these stats are a greater tool then watching a player play, but Holland, Babcock and company cant watch every player in the league for the entire game, every game of the year. When making a trade and/or signing a free agent, these stats should be considered.

Yes, they are and that's why watching the players is more important. I mean, things like intangibles, work ethic and grit can't be measured. Best example is Toews, doesn't score things of goals but he is easily a top 3 player behind Crosby and Stamkos << but to know that people have to watch the game instead of paying around with Excel sheets, calculators

I mean, GMs are already having enough information those new stats are just something to talk about fire the media.

[e_prime 1,936]

Right. ...because the stats guys aren't watching the games too.

How do you think the stats get collected? Computer algorithms?

...and don't get me wrong, I'm not saying that the "good ol' eyeball test" isn't good for anything...

Analyzing the game in different ways gives different perspectives and just furthers the possibilities in the game. It works both ways. Don't know why that's so hard for some to grasp.

[kliq 3,755]

Yes, they are and that's why watching the players is more important. I mean, things like intangibles, work ethic and grit can't be measured. Best example is Toews, doesn't score things of goals but he is easily a top 3 player behind Crosby and Stamkos << but to know that people have to watch the game instead of paying around with Excel sheets, calculators I mean, GMs are already having enough information those new stats are just something to talk about fire the media.

Nobody is saying advanced stats are MORE important then actually watching players, this is an argument you are having with nobody.

You can call the people that calculate advance statistics "nerds" and "math freaks" but at the end of the day you calling them names doesn't dismiss what they are doing. Advanced stats simply give you more information, and when handling a multi million (or even billion) dollar franchise, if you are a GM you should be using all the information/tools you can get.Baseball has been doing this for a while now, and some teams have had great success.

[PavelValerievichDatsyuk 1,935]

Yes, they are and that's why watching the players is more important. I mean, things like intangibles, work ethic and grit can't be measured. Best example is Toews, doesn't score things of goals but he is easily a top 3 player behind Crosby and Stamkos << but to know that people have to watch the game instead of paying around with Excel sheets, calculators I mean, GMs are already having enough information those new stats are just something to talk about fire the media.

Your Toews example doesn't help your point. The stat that you say is misleading for him is goals. That is not an advanced stat. The way you could demonstrate Toews is an elite player, despite not huge scoring totals, is through advanced stats. Sure, you can see it with the naked eye with Toews case, but there are often factors that can make a player look good when it is not necessarily due to them (think Ian White with Lidstrom, Bryzgalov with Pheonix). Or maybe factors may make a player look bad (say, if you're trying to judge whether Myers would be a worthy pickup). Advanced stats are an attempt to isolate those consideration.

[frankgrimes 1,836]

You can call the people that calculate advance statistics "nerds" and "math freaks" but at the end of the day you calling them names doesn't dismiss what they are doing. Advanced stats simply give you more information, and when handling a multi million (or even billion) dollar franchise, if you are a GM you should be using all the information/tools you can get.Baseball has been doing this for a while now, and some teams have had great success.

Actually I complemented them excel freaks and math geniuses is a positive. Baseball is a different much slower game and personally I highly believe GMs are having access to a lot of information.

I. Think Burke said it best :,, show me a stat that gives us an advantage and we are going to pay you a lot of money.. " but so far nobody has done that

[The Greek 323]

Actually I complemented them excel freaks and math geniuses is a positive. Baseball is a different much slower game and personally I highly believe GMs are having access to a lot of information. I. Think Burke said it best :,, show me a stat that gives us an advantage and we are going to pay you a lot of money.. " but so far nobody has done that

So you'd trust burke's opinion over babcock's?

[frankgrimes 1,836]

So you'd trust burke's opinion over babcock's?

Nope never but he is not wrong here. If a specific or specific stats can push your own team over the top for sure you are using them, but so far nobody has been able to produce such stats and I guess that's part of why some teams are hiring math geniuses.

Edited November 28, 2014 by frankgrimes