HFBoards

Go Back   HFBoards > NHL Eastern Conference > Atlantic Division > Montreal Canadiens
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Do advanced stats really tell the story?

Reply
 
Thread Tools
Old
08-27-2013, 12:22 PM
  #1
Lafontaine
Registered User
 
Join Date: May 2002
Posts: 377
vCash: 500
Do advanced stats really tell the story?

I've heard a bit about this in reviewing some recent threads. In the name of not wanting to be a (insert expletive), I want to give the notion a chance.

This might be a bit much to ask, but can anyone tell me who, statistically, were the worst Habs during play-offs last year? This would give me a decent idea if the numbers are worthless. For me, as a viewer, it was blatantly obvious who was contributing and who was not, so it seems an ideal sample for cross-referencing.

Lafontaine is offline   Reply With Quote
Old
08-29-2013, 04:02 AM
  #2
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 29,856
vCash: 925
Well, it depends a little on what numbers you put value on but of players who played most of the games, Gorgers, Diaz, Ryder.. Plekc too but some of that could be pointed to his role.. MaxPac compared to reg. season.

Little related, anyone know where one could find fenwick close for playoffs?

InjuredChoker is offline   Reply With Quote
Old
08-29-2013, 07:25 AM
  #3
sharks9
Registered User
 
sharks9's Avatar
 
Join Date: Jan 2012
Location: Canada
Posts: 13,045
vCash: 500
Looking at Corsi, Gorges, Moen, Diaz and Ryder were terrible.

Gallagher, Galchenyuk, Bouillon, Prust and Subban all had good Corsi.

sharks9 is offline   Reply With Quote
Old
08-29-2013, 07:30 AM
  #4
ECWHSWI
bought a MB jersey
 
ECWHSWI's Avatar
 
Join Date: Oct 2006
Posts: 20,495
vCash: 500
Quote:
Originally Posted by sharks9 View Post
Looking at Corsi, Gorges, Moen, Diaz and Ryder were terrible.

Gallagher, Galchenyuk, Bouillon, Prust and Subban all had good Corsi.
your post is a perfect example of why watching games will always be better than looking at (micro) stats.

ECWHSWI is offline   Reply With Quote
Old
08-29-2013, 07:55 AM
  #5
DAChampion
Registered User
 
DAChampion's Avatar
 
Join Date: May 2011
Location: Canberra, Australia
Country: Australia
Posts: 13,513
vCash: 500
I'll remind the stats lovers that it took computer programmers decades of work, among hundreds of talented and brilliant individuals, to develop software that could manage that game better than the most effective humans.

The statistical Gods of hockey are less intelligent than those computer programmers, less educated, less numerous, they have put in less effort, and they are analysing a more complex system with a greater number of mutually-interacting moving parts.

We should think of stats as a supplement (a good one!) to our understanding from watching the game, not as a replacement.

DAChampion is online now   Reply With Quote
Old
08-29-2013, 08:20 AM
  #6
MXD
Registered User
 
MXD's Avatar
 
Join Date: Oct 2005
Location: Hôlle
Posts: 30,943
vCash: 500
Quote:
Originally Posted by sharks9 View Post
Looking at Corsi, Gorges, Moen, Diaz and Ryder were terrible.

Gallagher, Galchenyuk, Bouillon, Prust and Subban all had good Corsi.
Looking the games pretty much gave the same pictures.

MXD is offline   Reply With Quote
Old
08-29-2013, 08:32 AM
  #7
coolasprICE
Registered User
 
coolasprICE's Avatar
 
Join Date: Mar 2008
Location: Montreal
Country: Canada
Posts: 9,832
vCash: 500
Quote:
Originally Posted by ECWHSWI View Post
your post is a perfect example of why watching games will always be better than looking at (micro) stats.
huh???

Or maybe this post is a good example why it's better to look at stats.

Gorges, Moen were crap. Diaz was overrated outside of his good start. Ryder wasn't terribad but he was streaky.

coolasprICE is offline   Reply With Quote
Old
08-29-2013, 08:53 AM
  #8
MXD
Registered User
 
MXD's Avatar
 
Join Date: Oct 2005
Location: Hôlle
Posts: 30,943
vCash: 500
I think the op should put (Playoffs) in parenthesis...

MXD is offline   Reply With Quote
Old
08-29-2013, 08:54 AM
  #9
MXD
Registered User
 
MXD's Avatar
 
Join Date: Oct 2005
Location: Hôlle
Posts: 30,943
vCash: 500
Quote:
Originally Posted by coolasprICE View Post
huh???

Or maybe this post is a good example why it's better to look at stats.

Gorges, Moen were crap. Diaz was overrated outside of his good start. Ryder wasn't terribad but he was streaky.
And they both really sucked in the playoffs.

MXD is offline   Reply With Quote
Old
08-29-2013, 08:58 AM
  #10
Whitesnake
Here we go again!
 
Whitesnake's Avatar
 
Join Date: Jan 2003
Country: Canada
Posts: 62,355
vCash: 725
Quote:
Originally Posted by ECWHSWI View Post
your post is a perfect example of why watching games will always be better than looking at (micro) stats.
Absolutely. Stats helps you understanding more about all the factors in a game and is there to add to an analysis...but people who simply uses that to explain how great or bad a player is are just plain wrong. But it's much more easier to do though than watching games and having your own opinion....

Whitesnake is offline   Reply With Quote
Old
08-29-2013, 08:59 AM
  #11
Dr Gonzo
#1 Jan Bulis Fan
 
Dr Gonzo's Avatar
 
Join Date: Dec 2009
Location: Bat Country
Posts: 4,355
vCash: 500
Quote:
Originally Posted by DAChampion View Post
I'll remind the stats lovers that it took computer programmers decades of work, among hundreds of talented and brilliant individuals, to develop software that could manage that game better than the most effective humans.

The statistical Gods of hockey are less intelligent than those computer programmers, less educated, less numerous, they have put in less effort, and they are analysing a more complex system with a greater number of mutually-interacting moving parts.

We should think of stats as a supplement (a good one!) to our understanding from watching the game, not as a replacement.
I don't think a single advanced stats user ever claimed it's the only tool to use when evaluating players or teams.

They'll all tell you that they use them with game viewership.

Except a certain advanced stats user faction: http://www.habseyesontheprize.com/20...t-watch-hockey



Last edited by Dr Gonzo: 08-29-2013 at 10:35 AM.
Dr Gonzo is offline   Reply With Quote
Old
08-29-2013, 09:01 AM
  #12
Whitesnake
Here we go again!
 
Whitesnake's Avatar
 
Join Date: Jan 2003
Country: Canada
Posts: 62,355
vCash: 725
Quote:
Originally Posted by coolasprICE View Post
huh???

Or maybe this post is a good example why it's better to look at stats.

Gorges, Moen were crap. Diaz was overrated outside of his good start. Ryder wasn't terribad but he was streaky.
Thing is....you could have also determined that by LOOKING at games.....Come on. I rarely look at those stats as we have experts around here who keeps doing it and I'm not into this and yet....I kept bashing Moen all year long. And said that it had to be Gorges worst season ever. A you can't determine a streaky Ryder by watching him play?

And then all the great ones from Gallagher to Galchenyuk, Subban and Co...wasn't it obvious of their good play by just watching him? I guess that the key guy here is Bouillon...yet, who here wants to re-sign him 4 years because he had a great Corsi?

Whitesnake is offline   Reply With Quote
Old
08-29-2013, 09:02 AM
  #13
Dr Gonzo
#1 Jan Bulis Fan
 
Dr Gonzo's Avatar
 
Join Date: Dec 2009
Location: Bat Country
Posts: 4,355
vCash: 500
Quote:
Originally Posted by Whitesnake View Post
Thing is....you could have also determined that by LOOKING at games....
Or you could do both and have the statistical proof as opposed to just a subjective opinion.

Cross reference both tools (eyes and stats) and you get a clearer picture of the truth. Which is what advanced stats users do. They watch the games and check the stats. It's not a bad thing.

Dr Gonzo is offline   Reply With Quote
Old
08-29-2013, 09:21 AM
  #14
Ohashi_Jouzu
Registered User
 
Ohashi_Jouzu's Avatar
 
Join Date: Apr 2007
Location: Halifax
Country: Japan
Posts: 28,447
vCash: 500
Quote:
Originally Posted by sharks9 View Post
Looking at Corsi, Gorges, Moen, Diaz and Ryder were terrible.

Gallagher, Galchenyuk, Bouillon, Prust and Subban all had good Corsi.
To flesh that out a little bit:

Those who faced the toughest competition (rel. Qual. Comp.) ended up with the worst Corsi scores. This would be expected over a short sample size, for sure (face better players, generate less possession stats). When you look at the GF/GA On/Off, it's also evident that the players with the best Corsi scores were also on the ice for more goals against than goals for - even relative to the rest of the team (even the guys with low Corsi scores).

Gallagher, for example, was on the ice for an average of 7.05 GA/60 minutes of ice time, while the team only allowed an average of 2.15 GA/60 minutes when he was off. Prust, White, Galchenyuk and Bouillon follow the same trend. Generated lots of chances, but were simply on the ice for too many goals against. Subban and Markov were, unsurprisingly, the only defensemen who were on the ice for proportionately much fewer goals against than the rest of the d-corps. Diaz was almost neutral (3.87 GA/60 on vs 3.21 GA/60 off) while facing the toughest competition of any defenseman (Corsi QoC/Rel QoC suggest Diaz and Gorges were put out to shut down top opposition, which we noticed, and I think most agree that perhaps that wasn't for the best).

An interesting case is Ryder, who, despite his negative Corsi, was on the ice for proportionately more goals for than the team scored while he was off, and was on the ice for fewer goals against than were allowed when he was off. He, Halpern, and Desharnais (if you can believe it) can also be judged as having generated the most goals from their defense, as they have among the highest On-ice SV% on the team but also generated some of the higher On-ice Shooting%, giving them a PDO >1000.

So, if I hadn't watched the games, I'd assume that those guys with good Corsi scores were doing lots of things to generate possession/shots, but didn't have great success beating the opposing goalie and/or didn't help out defensively perhaps as much as the team needed. Having watched the games, we know that some defensive lapses (and matchups) and a less than impressive Price probably figured into the GA as much as anything over the short sample, and we know that our offensive players were stymied by a hot goaltender.

Ohashi_Jouzu is offline   Reply With Quote
Old
08-29-2013, 09:24 AM
  #15
coolasprICE
Registered User
 
coolasprICE's Avatar
 
Join Date: Mar 2008
Location: Montreal
Country: Canada
Posts: 9,832
vCash: 500
Quote:
Originally Posted by Whitesnake View Post
Thing is....you could have also determined that by LOOKING at games.....Come on.
But read ECWHSWI post:

Quote:
your post is a perfect example of why watching games will always be better than looking at (micro) stats.
Not everyone has the ability to draw those conclusions.

coolasprICE is offline   Reply With Quote
Old
08-29-2013, 09:25 AM
  #16
MXD
Registered User
 
MXD's Avatar
 
Join Date: Oct 2005
Location: Hôlle
Posts: 30,943
vCash: 500
Quote:
Originally Posted by Ohashi_Jouzu View Post

An interesting case is Ryder, who, despite his negative Corsi, was on the ice for proportionately more goals for than the team scored while he was off, and was on the ice for fewer goals against than were allowed when he was off. He, Halpern, and Desharnais (if you can believe it) can also be judged as having generated the most goals from their defense, as they have among the highest On-ice SV% on the team but also generated some of the higher On-ice Shooting%, giving them a PDO >1000.
What advanced stats don't tell you is that both players looked completely loss on the ice.

MXD is offline   Reply With Quote
Old
08-29-2013, 09:58 AM
  #17
sharks9
Registered User
 
sharks9's Avatar
 
Join Date: Jan 2012
Location: Canada
Posts: 13,045
vCash: 500
Quote:
Originally Posted by ECWHSWI View Post
your post is a perfect example of why watching games will always be better than looking at (micro) stats.
What?

He asked who looked bad from an advanced stats point of view in the playoffs and I answered him.

Watching games is good but watching + using advanced stats is better.

sharks9 is offline   Reply With Quote
Old
08-29-2013, 10:04 AM
  #18
Ohashi_Jouzu
Registered User
 
Ohashi_Jouzu's Avatar
 
Join Date: Apr 2007
Location: Halifax
Country: Japan
Posts: 28,447
vCash: 500
Quote:
Originally Posted by sharks9 View Post
What?

He asked who looked bad from an advanced stats point of view in the playoffs and I answered him.

Watching games is good but watching + using advanced stats is better.
While I love stats (and especially advanced stats), the biggest problem is when people go from talking about proportion of total chances a player was involved in to the quality of those chances via the stats alone. Only watching the games gives any "real" idea of quality (which is where the notion of "skill" becomes most apparent), while stats obviously covers the quantity aspect (which are perhaps best used to compliment an "effectiveness" discussion, rather than "skill").

Ohashi_Jouzu is offline   Reply With Quote
Old
08-29-2013, 10:06 AM
  #19
MXD
Registered User
 
MXD's Avatar
 
Join Date: Oct 2005
Location: Hôlle
Posts: 30,943
vCash: 500
Quote:
Originally Posted by sharks9 View Post
What?

He asked who looked bad from an advanced stats point of view in the playoffs and I answered him.

Watching games is good but watching + using advanced stats is better.
Yeah, but by advanced stats, Desharnais was good (... well, was DECENT) in the playoffs. He actually was atrocious, looked lost and probably only outplayed Ryder.

Advanced stats can't seriously make DD better than he was. Mainly because "looking lost" isn't a stat column.

The same applies for Eller in his two regular seasons; the guy OFTEN looked lost -- good thing he appears to have turned the corner in that regard last season.

MXD is offline   Reply With Quote
Old
08-29-2013, 10:16 AM
  #20
Andy
Moderator
 
Andy's Avatar
 
Join Date: Jun 2008
Location: Montreal
Country: Canada
Posts: 25,261
vCash: 500
No they don't, but neither does subjective observation. That's why they should be used to complement each other.

Andy is offline   Reply With Quote
Old
08-29-2013, 10:35 AM
  #21
Fozz
Registered User
 
Fozz's Avatar
 
Join Date: Aug 2002
Location: Ottawa
Posts: 7,239
vCash: 500
Stats only tell a part of the story and will never grasp all elements of the game. People that base their opinions solely on stats only have a portion of the story.

Fozz is offline   Reply With Quote
Old
08-29-2013, 10:41 AM
  #22
Habsterix*
 
Join Date: Jul 2006
Location: BC
Country: Canada
Posts: 13,473
vCash: 500
Quote:
Do advanced stats really tell the story?
Simply put, no. Hockey is a team sport and it's simply trying to single players who play on a line. One example is when you hear that player so-and-so is playing sheltered minutes. In reality, it could very well be because of his linemate(s) that he's utilized the way he is. It's flawed but some swear by it... when it suits their argument.

Habsterix* is offline   Reply With Quote
Old
08-29-2013, 10:50 AM
  #23
Et le But
Registered User
 
Et le But's Avatar
 
Join Date: Nov 2010
Location: New York
Country: Argentina
Posts: 18,982
vCash: 500
Quote:
Originally Posted by DAChampion View Post
We should think of stats as a supplement (a good one!) to our understanding from watching the game, not as a replacement.
Has there ever been anyone who actually suggests that stats can replace seriously watching games? Even the most heavily stat-focused like MathMan have never gone that far.

I don't think that even in baseball stats will ever completely replace the qualitative analysis of watching a game. Like in everything else, it's about finding the right mix, and overcoming the cliches that the sports media tries to sell us.

Et le But is offline   Reply With Quote
Old
08-29-2013, 10:53 AM
  #24
Habsterix*
 
Join Date: Jul 2006
Location: BC
Country: Canada
Posts: 13,473
vCash: 500
Quote:
Originally Posted by Et le But View Post
Has there ever been anyone who actually suggests that stats can replace seriously watching games? Even the most heavily stat-focused like MathMan have never gone that far.

I don't think that even in baseball stats will ever completely replace the qualitative analysis of watching a game. Like in everything else, it's about finding the right mix, and overcoming the cliches that the sports media tries to sell us.
A smart man can make numbers speak for the point they are trying to make. Stats are no different. It's easy to take bits and pieces to support a point you're trying to make while ignoring others, but presenting it as purely factual. Stats (especially advanced stats) are flawed that way. At least that's my opinion.

Habsterix* is offline   Reply With Quote
Old
08-29-2013, 10:55 AM
  #25
habsfanatics*
 
Join Date: May 2012
Posts: 5,017
vCash: 500
Quote:
Originally Posted by Dr Gonzo View Post
I don't think a single advanced stats user ever claimed it's the only tool to use when evaluating players or teams.

They'll all tell you that they use them with game viewership.

Except a certain advanced stats user faction: http://www.habseyesontheprize.com/20...t-watch-hockey

Tell that to mathman who rates players as one of the worst in the entire league at their position based on advanced stats alone. I'm sure he watched tons of Sharks games and all 200 + dman. .


Last edited by habsfanatics*: 08-29-2013 at 11:00 AM.
habsfanatics* is offline   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 07:08 AM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"

vBulletin Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. ©2017 All Rights Reserved.