HFBoards

Go Back   HFBoards > NHL Western Conference > Pacific Division > Edmonton Oilers
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Oilers hire Tyler Dellow

Reply
 
Thread Tools
Old
08-13-2014, 02:55 PM
  #526
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Moose Coleman View Post
For the sake of argument, let's say you're right and not at all coming at this from a place of bias (a bias you admit and demonstrate whenever this subject comes up, which, if this were a court of law, would make you something of an unreliable witness): what's the relevance of this aspect of the discussions that took place in 2005 and before? It seems pretty obvious that the stats community has grown and evolved (Vic doesn't even post anymore), so beyond lingering personal animosity, what's the point in mentioning it?
The point was a blogger that posted unequivocally that such claims were never made, on the basis of said blogger exploring this debate and dialog indepth over the past year or so.

Seems a little odd to make one claim while at the same time admitting not to have followed the debate much longer than that.

That was the Sean Mcindoe piece I'm citing btw.

But agreed that one way or the other it doesn't have much relevance. just that Sean seems to be making a claim that no such inanity was ever involved or stated. Albeit in his come lately to the party zeal he's fine with labelling detractors as being solely disingenuous. That's what I take from the piece anyway.

He's establishing a narrative he's unaware of and seeming to claim that only the stats geeks are approaching the discussion honestly or with any substance. Seems onesided.

Replacement is offline   Reply With Quote
Old
08-13-2014, 03:17 PM
  #527
Psycho Dad
Oil Kings
 
Psycho Dad's Avatar
 
Join Date: Feb 2007
Location: Sherwood Park
Posts: 10,726
vCash: 610
Quote:
Originally Posted by Moose Coleman View Post
For the sake of argument, let's say you're right and not at all coming at this from a place of bias (a bias you admit and demonstrate whenever this subject comes up, which, if this were a court of law, would make you something of an unreliable witness): what's the relevance of this aspect of the discussions that took place in 2005 and before? It seems pretty obvious that the stats community has grown and evolved (Vic doesn't even post anymore), so beyond lingering personal animosity, what's the point in mentioning it?
I see a good part of the stats community gravitating to event-based observations and compiling actual data instead of trying to blend a bunch of Corsi-esque stats into a frothy exclamation of one player's worth. And that is commendable. Those folks have brought their data-gathering technique inline with observation-based valuation, by watching the games and recording. I wouldn't call it an evolution, more of a new beginning, if you will.

There are those who are stuck in their old ways, manipulating data to reach their desired conclusions. These people are claiming victory for their own methods, unaware that the whole movement has left them behind.

Then there are the stats fans, who I'm hoping are in the former group, and not the latter. If you're in the former group, honestly, we have no quarrel. Honestly, I'm right there with you.

Let's bring the new stats and the "seen 'em good" together. They're like spaghetti and meatballs. They need each other. Truce? Coalition?

*extends hand for handshake*

Psycho Dad is offline   Reply With Quote
Old
08-13-2014, 03:55 PM
  #528
Moose Coleman
Registered User
 
Moose Coleman's Avatar
 
Join Date: Apr 2012
Posts: 4,016
vCash: 500
Quote:
Originally Posted by Replacement View Post
The point was a blogger that posted unequivocally that such claims were never made, on the basis of said blogger exploring this debate and dialog indepth over the past year or so.

Seems a little odd to make one claim while at the same time admitting not to have followed the debate much longer than that.

That was the Sean Mcindoe piece I'm citing btw.
See, this is what I mean about you distorting things. Here's the exact quote:

Quote:
That still seems to surprise some people. Iíve had conversations with more analytics folks than I can count over the last year, including most of the biggest names. Iíve followed the stats blogs. Iíve read as many of the reports as I could get my hands on, then read them again until I could start to understand them. Iíve asked questions, many of them hopelessly dumb, in an attempt to get my head around this stuff.

And in all that time, Iíve literally never once encountered anyone who thinks that stats tell the whole story
That's a pretty straightforward description of his own subjective experience. Why you feel compelled to use your own subjective experience to assail this guy is unclear.

Quote:
But agreed that one way or the other it doesn't have much relevance. just that Sean seems to be making a claim that no such inanity was ever involved or stated. Albeit in his come lately to the party zeal he's fine with labelling detractors as being solely disingenuous.
He stated he's never heard anyone say stats tell the whole story.

That is not the same as claiming no one in the history of this debate has said stats tell the whole story.

Quote:
That's what I take from the piece anyway.
Uh huh. That says more about your own perspective/biases than anything else.

Quote:
He's establishing a narrative he's unaware of and seeming to claim that only the stats geeks are approaching the discussion honestly or with any substance. Seems onesided.
So what narrative are you trying to establish by misrepresenting what he wrote?

Sounds to me like you're nursing a festering wound from those bygone days and projecting your issues with the stats guys of yore onto today's believers. But that's just my read.

Moose Coleman is offline   Reply With Quote
Old
08-13-2014, 04:08 PM
  #529
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Moose Coleman View Post
See, this is what I mean about you distorting things. Here's the exact quote:



That's a pretty straightforward description of his own subjective experience. Why you feel compelled to use your own subjective experience to assail this guy is unclear.



He stated he's never heard anyone say stats tell the whole story.

That is not the same as claiming no one in the history of this debate has said stats tell the whole story.



Uh huh. That says more about your own perspective/biases than anything else.



So what narrative are you trying to establish by misrepresenting what he wrote?

Sounds to me like you're nursing a festering wound from those bygone days and projecting your issues with the stats guys of yore onto today's believers. But that's just my read.
WE all establish narratives. I've read this and the writers other blogs. We'll agree to disagree for the sake of brevity. This guy is as much in a camp as any of us.

I would say that this is stated unequivocally, apparently you disagree;

Quote:
The stat guy with all the answers and no time for old-fashioned observation is a straw man. He exists only in the realm of anti-stat fantasy. Let him go.
The crux of the above being whether it was made up strawman. Was it? Really?


Tend to agree with Psychodad that dialog can possibly move forward and that spaghetti and meatballs sounds fine. But now the arguing on who's going to be meatballs..

Replacement is offline   Reply With Quote
Old
08-13-2014, 04:12 PM
  #530
Moose Coleman
Registered User
 
Moose Coleman's Avatar
 
Join Date: Apr 2012
Posts: 4,016
vCash: 500
Quote:
Originally Posted by Replacement View Post
WE all establish narratives. I've read this and the writers other blogs. We'll agree to disagree for the sake of brevity. This guy is as much in a camp as any of us.
Unbelievable.

I guess I'm done with this, not because I think rapprochement is needed, but because this kind of bald-faced prevarication is simply too much.

Moose Coleman is offline   Reply With Quote
Old
08-13-2014, 04:13 PM
  #531
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Moose Coleman View Post
Unbelievable.

I guess I'm done with this, not because I think rapprochement is needed, but because this kind of bald-faced prevarication is simply too much.
Did you miss the quote above that was unequivocal from the blogger? Apparently you did.

What more do you require?

Replacement is offline   Reply With Quote
Old
08-13-2014, 04:19 PM
  #532
Moose Coleman
Registered User
 
Moose Coleman's Avatar
 
Join Date: Apr 2012
Posts: 4,016
vCash: 500
Quote:
Originally Posted by Replacement View Post
Did you miss the quote above that was unequivocal from the blogger? Apparently you did.

What more do you require?
Just saw that as you made the edit while I was posting. Changes nothing, really.

Quote:
The stat guy with all the answers and no time for old-fashioned observation is a straw man. He exists only in the realm of anti-stat fantasy. Let him go.
That ten years ago some stats guys claimed to be able to tell everything from the game sheet (according to your telling) to provoke (again, by your telling) in no way invalidates McIndoe's claim that such people don't exist today. Let's put it another way: when was the last time you heard anyone say stats tell the whole story (and actually mean it)?

Moose Coleman is offline   Reply With Quote
Old
08-13-2014, 04:32 PM
  #533
Master Lok
Registered User
 
Join Date: Jul 2003
Location: Edmonton
Posts: 6,901
vCash: 500
Quote:
Originally Posted by Replacement View Post
I didn't see much in Willis rebuttal that was convincing. Both sides aren't capturing the argument very well imo.
And what is the argument as you see it?

Master Lok is offline   Reply With Quote
Old
08-13-2014, 04:34 PM
  #534
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Moose Coleman View Post
Just saw that as you made the edit while I was posting. Changes nothing, really.
Oh of course not..

Quote:
That ten years ago some stats guys claimed to be able to tell everything from the game sheet (according to your telling) to provoke (again, by your telling) in no way invalidates McIndoe's claim that such people don't exist today. Let's put it another way: when was the last time you heard anyone say stats tell the whole story (and actually mean it)?
On this board? Around the last time people posting utter nonsense under multiple pseudonames were banned from the board. With much of the reason they were banned being undetectable from the posts currently on record that were left in place and not erased, deleted, scoured..

People used to have to submit screen captures (and did) to get some of these people booted.

Replacement is offline   Reply With Quote
Old
08-13-2014, 04:39 PM
  #535
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Master Lok View Post
And what is the argument as you see it?
That both sides are waging not very well done point counterpoint.

tbh I'm not convinced I'd want either on my side in a debate. just saying.

The one thing agreed on is the debate is pointless. Lets see what the results end up being.

Still waiting though to any link that's forecasting the 2014-15 season standings. predictive validity and all that.

Shouldn't be hard to find..

Replacement is offline   Reply With Quote
Old
08-13-2014, 04:50 PM
  #536
Moose Coleman
Registered User
 
Moose Coleman's Avatar
 
Join Date: Apr 2012
Posts: 4,016
vCash: 500
Quote:
Originally Posted by Replacement View Post
Oh of course not..

On this board? Around the last time people posting utter nonsense under multiple pseudonames were banned from the board. With much of the reason they were banned being undetectable from the posts currently on record that were left in place and not erased, deleted, scoured..

People used to have to submit screen captures (and did) to get some of these people booted.
And what year was this?

Moose Coleman is offline   Reply With Quote
Old
08-13-2014, 05:19 PM
  #537
Master Lok
Registered User
 
Join Date: Jul 2003
Location: Edmonton
Posts: 6,901
vCash: 500
Quote:
Originally Posted by Replacement View Post
That both sides are waging not very well done point counterpoint.

tbh I'm not convinced I'd want either on my side in a debate. just saying.

The one thing agreed on is the debate is pointless. Lets see what the results end up being.

Still waiting though to any link that's forecasting the 2014-15 season standings. predictive validity and all that.

Shouldn't be hard to find..
Agreed that the whole debate is pointless.

I was afraid that the debate was whether or not analytics was useless or a fad, as suggested by Derek Van Diest.

Master Lok is offline   Reply With Quote
Old
08-13-2014, 05:27 PM
  #538
Section337
Registered User
 
Join Date: Jul 2007
Location: Edmonton, AB
Posts: 4,285
vCash: 500
Quote:
Originally Posted by Replacement View Post
The writer states to have never met a stats geek that makes claims based on stats only.

Any of Igor, Vic Ferrari, several of the other Stats hangers on at the time say hi.

The trouble with such an assessment is it seems to miss what was going on around here and the blogosphere 5-10 yrs ago which many of us remember and with stats geeks routinely stating that they know more about Horcoff than a regular viewer just from running their scripts..

To have a community engage in such claims for years and then years later state it never happened...well....
His engagement with people may be different than other people's engagements with those people. He may have stayed out of some of the crazier discussions. Or talked to them directly, which has a better chance of understanding each sides position, than does a internet debate.

Section337 is offline   Reply With Quote
Old
08-13-2014, 10:35 PM
  #539
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Section337 View Post
His engagement with people may be different than other people's engagements with those people. He may have stayed out of some of the crazier discussions. Or talked to them directly, which has a better chance of understanding each sides position, than does a internet debate.
You mean this isn't the place to understand others thoughts and feelings?

I'm doing the internets wrong.

Replacement is offline   Reply With Quote
Old
08-14-2014, 03:46 PM
  #540
Valic
BOOOOOOOOOO
 
Valic's Avatar
 
Join Date: Jun 2007
Location: Canada
Country: Canada
Posts: 8,726
vCash: 500
Quote:
Originally Posted by topchowda View Post
Wow, this article is an absolute must read. Its like he wrote everything I was thinking. This is a perfect middle ground article marrying both sides of the argument.

This should be a prerequiste read before entering into a stats debate
Here's the thing about what you just said. It's not a middle ground at all. This is the stance of where most advanced stats guys sit.

Of course there are outliers, but the majority of the people doing the work, and those that actually follow the work at more than a glance sit very close this this view point.

Valic is offline   Reply With Quote
Old
08-18-2014, 06:00 AM
  #541
The Human Torch
Registered User
 
The Human Torch's Avatar
 
Join Date: Sep 2005
Location: Amsterdam
Country: Netherlands
Posts: 5,172
vCash: 500
Send a message via MSN to The Human Torch
I don't even know what this thread is discussing any longer.

The Human Torch is offline   Reply With Quote
Old
08-18-2014, 09:18 AM
  #542
J4M13M
Registered User
 
Join Date: Jun 2014
Location: Saint John NB
Country: Canada
Posts: 60
vCash: 500
I've been thining about this Dellow hiring a bit recently, and I see the stats thingas a smokescreen. The guy has spent hundreds of hours watching isolated videao andanalyzing minute portions of the game. His work on the Oilers' offensive zone faceoff alignment alone could be enough to get Yakupov 10 more goals (hyperbole maybe, but if we see the second line getting more sustained pressure off draws, it'll definitely help Yak's numbers).

I like the idea of having a database of stats, and I think Dellow is among the best at that. But he's also pretty good at finding evidence to prove his points on a lot of hockey-related matters. I think if the coaching staff is willing to study his input, they'll see lots of minor adjustments they can experiment with through the year.

J4M13M is offline   Reply With Quote
Old
08-18-2014, 12:48 PM
  #543
joestevens29
Registered User
 
joestevens29's Avatar
 
Join Date: Apr 2009
Posts: 25,767
vCash: 50
Quote:
Originally Posted by J4M13M View Post
I've been thining about this Dellow hiring a bit recently, and I see the stats thingas a smokescreen. The guy has spent hundreds of hours watching isolated videao andanalyzing minute portions of the game. His work on the Oilers' offensive zone faceoff alignment alone could be enough to get Yakupov 10 more goals (hyperbole maybe, but if we see the second line getting more sustained pressure off draws, it'll definitely help Yak's numbers).

I like the idea of having a database of stats, and I think Dellow is among the best at that. But he's also pretty good at finding evidence to prove his points on a lot of hockey-related matters. I think if the coaching staff is willing to study his input, they'll see lots of minor adjustments they can experiment with through the year.
Is his stats really going to tell us how to fix things or just that we need to fix things? I don't doubt Eakins will ask him about what's not working, but what does Dellow really know about what's going to work unless it's already happened?

It's as if we need to try 10 different thing in 20 different games and go from there, but we know that won't happen.

joestevens29 is offline   Reply With Quote
Old
08-18-2014, 01:35 PM
  #544
Beerfish
Registered User
 
Beerfish's Avatar
 
Join Date: Apr 2007
Posts: 11,769
vCash: 500
One just has to be aware enough about hockey stats to realize that they are like accounting and can easily be manipulate to fit your opinion or make you look like you are never wrong.

A very big key to this type of data for me is 'time period'. Let's look at a statement like 'Colorado can't sustain their performance because their numbers are unrealistic.' When comments were made about this was any time frame mentioned? If a guy makes that statement 1/2 way though the year and that team does not fall off the cliff but does so the next year he is right. They can make the same statement about the Leafs, in that case they did fall off the cliff.

Wait long enough without coming up with a time period and you will never be wrong.

Beerfish is offline   Reply With Quote
Old
08-18-2014, 01:41 PM
  #545
J4M13M
Registered User
 
Join Date: Jun 2014
Location: Saint John NB
Country: Canada
Posts: 60
vCash: 500
Quote:
Originally Posted by joestevens29 View Post
Is his stats really going to tell us how to fix things or just that we need to fix things? I don't doubt Eakins will ask him about what's not working, but what does Dellow really know about what's going to work unless it's already happened?

It's as if we need to try 10 different thing in 20 different games and go from there, but we know that won't happen.
His analysis on Oiler Offensive Zone faceoffs was not purely stats-based. It was based on a stat telling him something was broken, and he worked up a high wordcount on identifying the issue, and some adjustments that *might* fix it.

The answer is no, he doesn't know what will fix it, and no NHL coach in his right mind is going to take the word of a blogger on faith and alter his coaching philosophy anyway. But if you look at the combination of stats, video, and analysis from that work alone, you'll see how it might work. For me, that and the coach's comments on including him in the conversation are enough to give me some optimism.

J4M13M is offline   Reply With Quote
Old
08-18-2014, 02:41 PM
  #546
Dorian2
Positional Bias.
 
Dorian2's Avatar
 
Join Date: Jul 2009
Location: Edmonton
Country: Canada
Posts: 6,091
vCash: 50
Quote:
Originally Posted by Beerfish View Post
One just has to be aware enough about hockey stats to realize that they are like accounting and can easily be manipulate to fit your opinion or make you look like you are never wrong.

A very big key to this type of data for me is 'time period'. Let's look at a statement like 'Colorado can't sustain their performance because their numbers are unrealistic.' When comments were made about this was any time frame mentioned? If a guy makes that statement 1/2 way though the year and that team does not fall off the cliff but does so the next year he is right. They can make the same statement about the Leafs, in that case they did fall off the cliff.

Wait long enough without coming up with a time period and you will never be wrong.
Agree 100% with everything you've stated in this post.

Dorian2 is offline   Reply With Quote
Old
08-18-2014, 03:07 PM
  #547
Mr Sakich
Registered User
 
Mr Sakich's Avatar
 
Join Date: Mar 2002
Location: Motel 35
Posts: 8,100
vCash: 500
Quote:
Originally Posted by Replacement View Post
You mean this isn't the place to understand others thoughts and feelings?

I'm doing the internets wrong.
Replacement, you don't have to answer this question but I am curious. When you were a teenager, did your parents start to drink a lot?

Mr Sakich is offline   Reply With Quote
Old
08-18-2014, 06:39 PM
  #548
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Beerfish View Post
One just has to be aware enough about hockey stats to realize that they are like accounting and can easily be manipulate to fit your opinion or make you look like you are never wrong.

A very big key to this type of data for me is 'time period'. Let's look at a statement like 'Colorado can't sustain their performance because their numbers are unrealistic.' When comments were made about this was any time frame mentioned? If a guy makes that statement 1/2 way though the year and that team does not fall off the cliff but does so the next year he is right. They can make the same statement about the Leafs, in that case they did fall off the cliff.

Wait long enough without coming up with a time period and you will never be wrong.
Certain astrologists used to make this kind of statement like it was an artform..

Just saying.

I'm amazed as well what the advanced stats crowd considers as supporting evidence.

Again where is any indication of an attempt at predictive validity? Wheres the Corsicoligists proclamation of 14-15 standings. Lineups are established, team cors's should be a known. Where the predictions. Been asking this for several years.. I predict I'll continue to have to ask where these are.. see what I did there with the time frame..

Replacement is offline   Reply With Quote
Old
08-18-2014, 06:41 PM
  #549
Replacement
Now with 9% more zen
 
Replacement's Avatar
 
Join Date: Apr 2005
Location: Hockey Hell
Country: Canada
Posts: 38,068
vCash: 500
Quote:
Originally Posted by Mr Sakich View Post
Replacement, you don't have to answer this question but I am curious. When you were a teenager, did your parents start to drink a lot?
You have the cause and effect wrong. Or I do..

No, they started when I was young. Must've been my karma, thanks for asking..

Replacement is offline   Reply With Quote
Old
08-18-2014, 08:24 PM
  #550
topchowda
Registered User
 
topchowda's Avatar
 
Join Date: Aug 2012
Posts: 7,430
vCash: 500
Quote:
Originally Posted by Beerfish View Post
One just has to be aware enough about hockey stats to realize that they are like accounting and can easily be manipulate to fit your opinion or make you look like you are never wrong.

A very big key to this type of data for me is 'time period'. Let's look at a statement like 'Colorado can't sustain their performance because their numbers are unrealistic.' When comments were made about this was any time frame mentioned? If a guy makes that statement 1/2 way though the year and that team does not fall off the cliff but does so the next year he is right. They can make the same statement about the Leafs, in that case they did fall off the cliff.

Wait long enough without coming up with a time period and you will never be wrong.
When the time frame is within a 40 game period, its pretty much the same cycle. If you look at s% ranges, they typically only last for 1 season (82 games), or just over 100 if they extend into the next season. In the case of the Avalanche, nearly every person with a basic handle on analytics can/ could see their demise. And when they got fed in the playoffs, I think this is a major sign of whats to come. But in response to your statement, I think next season, 35 games in, will be the time frame. I will judge the season over the full 82 games, but I will wage that by game 35 the true Avalanche will show.

This is a great introductory read

http://www.arcticicehockey.com/2013/...ckey-analytics

Also really good links to various different subjects are contained within this article. I have used it alot over time

One of my favourite lines:

Quote:
Corsi and Fenwick, these are not the major causes for wins, but are the by-product of striving for what does create wins: scoring chances and puck possession. For this very reason they correlate to these things very, very strongly and therefore can be used as a proxy for the same thing
Also, the eye test can also be used to say anything you want, or to prove anything. The eye test, used solely, is not anymore effective then using stats purely. For example, before the reliance of analytics, many pro and amatuer scouts used the eye test to pick prospects. You would think that with having scouts with 25 years of experience, they would be right on the money with picks and evaluating pro players. But then how to you explain the numerous busts, the picks from the later rounds that turned out to be stars (no scout in their right mind would knowlingly let a future star drop to the later rounds), or the players acquired in trades or FA that turned out to be horrible? The answer is because scouts, like fans, have bias. They value what they perceive as good, and ignore other aspects. Some teams value size so pick prospects based on size and hope they develop skill. Advanced stats can step in and show how wrong this is. Or scouts can value defensive play, even if the player can not generate any offense. There is no single one number that will show a players true value, or be able to tell you how good he is. But by using a combination in the right contexts, and by applying the eye test to verify one another, you can get a MUCH better picture then by simply purely using the eye test, or simply using a stats based approach

The value of the eye test is pretty overblown. Yet in the future I am winngly to bet people will say "why isnt Fayne, or Stralman, or Perrault a star yet", or if a player who advanced stats people like goes on to be bad, people will put the blame solely on advanced stats, while ignoring the 100s of other busts the eye tests failed to see.


Last edited by topchowda: 08-18-2014 at 08:49 PM.
topchowda 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 09:44 AM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"
Contact Us - HFBoards - Archive - Privacy Statement - Terms of Use - Advertise - Top - AdChoices

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. ©2014 All Rights Reserved.