HFBoards

Go Back   HFBoards > Fantasy Hockey Talk > All Time Draft
All Time Draft Fantasy league where players of the past and present meet.

ATD 2012 - Draft Thread VI

Closed Thread
 
Thread Tools
Old
02-25-2012, 12:13 PM
  #251
BillyShoe1721
Terriers
 
BillyShoe1721's Avatar
 
Join Date: Mar 2007
Location: Philadelphia, PA
Country: United States
Posts: 16,640
vCash: 8400
Send a message via AIM to BillyShoe1721
Quote:
Originally Posted by arrbez View Post
I know this has been explained before, but I forget.

What specifically goes into these estimated TOI numbers for guys from the 60's, 70's, etc?
Hockey Reference has numbers for goals on ice for and against starting in 67-68, so before then, I don't know how people do it. The way I do it for those years and after is I add up the goals they were on the ice for and against and divide it by the number of games they played that season. Do it for every defenseman, and you've got an estimated TOI. You can do it with PP time and SH time as well.

BillyShoe1721 is offline  
Old
02-25-2012, 12:15 PM
  #252
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by BillyShoe1721 View Post
Hockey Reference has numbers for goals on ice for and against starting in 67-68, so before then, I don't know how people do it. The way I do it for those years and after is I add up the goals they were on the ice for and against and divide it by the number of games they played that season. Do it for every defenseman, and you've got an estimated TOI. You can do it with PP time and SH time as well.
This method will vastly overrate the ice times of offensive defensemen who are on ice for both more GF and GA, right? Or am I missing something?

TheDevilMadeMe is online now  
Old
02-25-2012, 12:17 PM
  #253
arrbez
bad chi
 
arrbez's Avatar
 
Join Date: Jun 2004
Location: Toronto
Country: Canada
Posts: 12,611
vCash: 500
Send a message via MSN to arrbez
Quote:
Originally Posted by BillyShoe1721 View Post
Hockey Reference has numbers for goals on ice for and against starting in 67-68, so before then, I don't know how people do it. The way I do it for those years and after is I add up the goals they were on the ice for and against and divide it by the number of games they played that season. Do it for every defenseman, and you've got an estimated TOI. You can do it with PP time and SH time as well.
Wouldn't that be skewed by style of play? I'd imagine Pavel Bure was on the ice for more goals in both directions than Guy Carbonneau was, as an example.

arrbez is offline  
Old
02-25-2012, 12:25 PM
  #254
BillyShoe1721
Terriers
 
BillyShoe1721's Avatar
 
Join Date: Mar 2007
Location: Philadelphia, PA
Country: United States
Posts: 16,640
vCash: 8400
Send a message via AIM to BillyShoe1721
It's a crude method, but do you see any other way of doing it? I've found that it is generally effective. For guys that have really high totals, like Phil Housley or Reed Larson, it may overrate their totals, but if they were playing alongside a guy that was clearly better, we put an asterisk next to it. It's not perfect, but it's better than nothing.

BillyShoe1721 is offline  
Old
02-25-2012, 12:27 PM
  #255
arrbez
bad chi
 
arrbez's Avatar
 
Join Date: Jun 2004
Location: Toronto
Country: Canada
Posts: 12,611
vCash: 500
Send a message via MSN to arrbez
Quote:
Originally Posted by BillyShoe1721 View Post
It's a crude method, but do you see any other way of doing it?
Haha, not at all . It's better than nothing, but as you say, it does need to be taken with a grain of salt.

arrbez is offline  
Old
02-25-2012, 12:31 PM
  #256
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
If the method was truly distorting ice times in a consistent fashion, then I'd say it is worse than nothing because it does nothing but cause distortion.

I've seen posts before that TOI estimates have a very high correlation with actual TOI in seasons we have it for, which is the only reason I pay attention to the estimates.

TheDevilMadeMe is online now  
Old
02-25-2012, 12:31 PM
  #257
Hawkey Town 18
Moderator
 
Hawkey Town 18's Avatar
 
Join Date: Jun 2009
Location: Chicago, IL
Posts: 4,159
vCash: 500
I'm pretty sure I've heard seventies say that when the estimated method was used for recent years where actual TOI numbers are available the estimated results were pretty close to the real numbers


EDIT: TDMM beat me to it!

Hawkey Town 18 is offline  
Old
02-25-2012, 12:32 PM
  #258
BillyShoe1721
Terriers
 
BillyShoe1721's Avatar
 
Join Date: Mar 2007
Location: Philadelphia, PA
Country: United States
Posts: 16,640
vCash: 8400
Send a message via AIM to BillyShoe1721
Seventies/overpass may have a different method for doing it. You can also remove power play goals to make it more even for defensive defensemen who tend to be underrated by this system. Or remove PP and PK goals against to get ES TOI.

BillyShoe1721 is offline  
Old
02-25-2012, 12:34 PM
  #259
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 23,629
vCash: 500
Quote:
Originally Posted by BillyShoe1721 View Post
Hockey Reference has numbers for goals on ice for and against starting in 67-68, so before then, I don't know how people do it. The way I do it for those years and after is I add up the goals they were on the ice for and against and divide it by the number of games they played that season. Do it for every defenseman, and you've got an estimated TOI. You can do it with PP time and SH time as well.
First - no one has ever tried to approximate icetime before expansion. There is no data to use for that.

Second - why would you waste your time doing the above to get inaccurate results, when you can just use the file, which is based on a much more sophisticated and accurate formula?

TDMM - I used to think that it could overstate the toi of high risk/high reward players, but the results don't seem to bear that out. There are well known offensive specialists who come out low just like you'd expect, while a guy like Dave Burrows comes out high like you'd expect.

Arrbez - pavel bure played more minutes than carbonneau.

seventieslord is online now  
Old
02-25-2012, 12:35 PM
  #260
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 23,629
vCash: 500
Quote:
Originally Posted by BillyShoe1721 View Post
Seventies/overpass may have a different method for doing it. You can also remove power play goals to make it more even for defensive defensemen who tend to be underrated by this system. Or remove PP and PK goals against to get ES TOI.
It is not our method. It was done by people more advanced than us. I highly recommend it over what you are doing.

seventieslord is online now  
Old
02-25-2012, 12:39 PM
  #261
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 2,982
vCash: 500
Quote:
Originally Posted by Sturminator View Post
I will add that I have never really bought this argument when used for Derian Hatcher, either. Maybe for the first few years of his prime, but certainly not for the latter part. Those Stars teams were consistently among the league's best, and Hatcher got a ton of exposure during that period. I think the "AST voting underrates him" argument is legit for defensive defensemen on bad teams, but otherwise, it is a pretty weak argument.

Kevin Lowe is actually a perfect example of a good team defensive defenseman who did get AST credit for what he was doing. I mean...Lowe was a fine defenseman, but I would be hard-pressed to call him better than the voting numbers indicate. The best players at all positions tend to be effective going both ways. There is no doubt that, since expansion, pure offensive defensemen have fared better in AST voting than pure defensive defensemen - this is a clear distortion, and I do not believe, for example, that Phil Housley was really a better hockey player than Kevin Lowe.

But I don't believe, at all, that defensive defensemen have been underrated in comparison to guys who were two-way players. When this line of thinking gets to the point of aggrandizing a guy like McCrimmon in comparison to a guy like Beck (who was good for maybe 10 points/season more than McCrimmon, on average), I think it has outlived its usefulness.
Defensive defensemens ASTs vote behind the top-4 is an obscure stat to use at best. Counting stray votes as something significant is laughable. Should we really acknowledge Prongers 2 votes in '97? J.J. Daigneault got a vote too he must be tremendous.

When will people start to realise that after the elite the votes are pretty randomized, petty and in fact probably full with homerism. I wouldn't exactly brag about McCrimmon getting four votes earning him the 12th place. Desjardins had one vote in 01 placing him 13th.

Hobnobs is offline  
Old
02-25-2012, 12:39 PM
  #262
arrbez
bad chi
 
arrbez's Avatar
 
Join Date: Jun 2004
Location: Toronto
Country: Canada
Posts: 12,611
vCash: 500
Send a message via MSN to arrbez
Quote:
Originally Posted by seventieslord View Post
Arrbez - pavel bure played more minutes than carbonneau.
No doubt, but those were just random examples of risky vs. conservative players. My point is that if Pavel Bure and Guy Carbonneau played an entire season at the exact same ES minutes, you can guarantee that more goals would be scored (for and against) with Bure on the ice than Carbonneau. And that would apply to all players at varying degrees.

arrbez is offline  
Old
02-25-2012, 12:41 PM
  #263
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by Hobnobs View Post
Defensive defensemens ASTs vote behind the top-4 is an obscure stat to use at best. Counting stray votes as something significant is laughable. Should we really acknowledge Prongers 2 votes in '97? J.J. Daigneault got a vote too he must be tremendous.

When will people start to realise that after the elite the votes are pretty randomized, petty and in fact probably full with homerism. I wouldn't exactly brag about McCrimmon getting four votes earning him the 12th place. Desjardins had one vote in 01 placing him 13th.
Wouldn't the homer votes cancel each other out? And nobody who does these things responsibly (which includes Sturm, 70s, and me) counts a season when a player only got one vote.

TheDevilMadeMe is online now  
Old
02-25-2012, 12:42 PM
  #264
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 23,629
vCash: 500
Quote:
Originally Posted by Hobnobs View Post
Defensive defensemens ASTs vote behind the top-4 is an obscure stat to use at best. Counting stray votes as something significant is laughable. Should we really acknowledge Prongers 2 votes in '97? J.J. Daigneault got a vote too he must be tremendous.

When will people start to realise that after the elite the votes are pretty randomized, petty and in fact probably full with homerism. I wouldn't exactly brag about McCrimmon getting four votes earning him the 12th place. Desjardins had one vote in 01 placing him 13th.
Nobody here, except Dreakmur, has ever tried to say one vote is significant, and I'm pretty sure he got shamed out of it last draft with the Ted Green scandal.

There is actually good depth in the 60s-80s in the allstar voting. Take a look at the results. Far more than just the top-4 are significant and useful.

seventieslord is online now  
Old
02-25-2012, 12:44 PM
  #265
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 23,629
vCash: 500
Quote:
Originally Posted by arrbez View Post
No doubt, but those were just random examples of risky vs. conservative players. My point is that if Pavel Bure and Guy Carbonneau played an entire season at the exact same ES minutes, you guarantee that more goals would be scored (for and against) with Bure on the ice than Carbonneau. And that would apply to all players at varying degrees.
I invited Iain fyffe to explain this better than I can, but I think there is a part to this formula that washes this out, not entirely, but at least partially.

seventieslord is online now  
Old
02-25-2012, 12:51 PM
  #266
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by seventieslord View Post
First - no one has ever tried to approximate icetime before expansion. There is no data to use for that.

Second - why would you waste your time doing the above to get inaccurate results, when you can just use the file, which is based on a much more sophisticated and accurate formula?

TDMM - I used to think that it could overstate the toi of high risk/high reward players, but the results don't seem to bear that out. There are well known offensive specialists who come out low just like you'd expect, while a guy like Dave Burrows comes out high like you'd expect.

Arrbez - pavel bure played more minutes than carbonneau.
I asked in the Hockey Summary Project stickied thread on the HOH board, but apparently you didn't see it. Is there any way we can include a link to the files in one of the stickied threads on the HOH board, or would that overwhelm whoever's drop box they are in?

TheDevilMadeMe is online now  
Old
02-25-2012, 12:58 PM
  #267
Nalyd Psycho
Registered User
 
Nalyd Psycho's Avatar
 
Join Date: Feb 2002
Location: No Bandwagon
Country: Canada
Posts: 22,718
vCash: 500
Quote:
Originally Posted by Nalyd Psycho View Post
The Minnesota Fighting Saints are pleased to select, from Karlskoga Sweden, Bengt-Åke Gustafsson, RW
To expand on this, way back when I was looking at selke votes for RWs, I found only three names got noteworthy support. Kurri, Lehtinen and Gustafsson. Lehtinen of course played both wings and Gustafsson seemingly played all three forward positions. But those three were the only players to take regular shifts at RW and get regular support from Selke voters.

__________________
Every post comes with the Nalyd Psycho Seal of Approval.
Nalyd Psycho is offline  
Old
02-25-2012, 01:11 PM
  #268
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 2,982
vCash: 500
Quote:
Originally Posted by seventieslord View Post
Nobody here, except Dreakmur, has ever tried to say one vote is significant, and I'm pretty sure he got shamed out of it last draft with the Ted Green scandal.

There is actually good depth in the 60s-80s in the allstar voting. Take a look at the results. Far more than just the top-4 are significant and useful.
there might be depth but only to a point where there is bunch of guys with 7-1 points.



Quote:
Wouldn't the homer votes cancel each other out? And nobody who does these things responsibly (which includes Sturm, 70s, and me) counts a season when a player only got one vote.
Sturm just added a bunch of 4 votes and downwards in hist list a few posts ago. Like Harpers 17th place matters....

Hobnobs is offline  
Old
02-25-2012, 01:16 PM
  #269
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by Hobnobs View Post


Sturm just added a bunch of 4 votes and downwards in hist list a few posts ago. Like Harpers 17th place matters....
Is there a GM here who doesn't realize that a 17th place finish is going to only be a few votes?

TheDevilMadeMe is online now  
Old
02-25-2012, 01:21 PM
  #270
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 23,629
vCash: 500
Quote:
Originally Posted by TheDevilMadeMe View Post
I asked in the Hockey Summary Project stickied thread on the HOH board, but apparently you didn't see it. Is there any way we can include a link to the files in one of the stickied threads on the HOH board, or would that overwhelm whoever's drop box they are in?
I can do that. I guess the links to public dropbox folders don't last forever so i'd have to keep refreshing it., but it's no big deal.

I don't think the demand for this type of thing is so huge that it would overwhelm my account, but I guess we'll see.

seventieslord is online now  
Old
02-25-2012, 01:22 PM
  #271
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by Sturminator View Post
I will add that I have never really bought this argument when used for Derian Hatcher, either. Maybe for the first few years of his prime, but certainly not for the latter part. Those Stars teams were consistently among the league's best, and Hatcher got a ton of exposure during that period. I think the "AST voting underrates him" argument is legit for defensive defensemen on bad teams, but otherwise, it is a pretty weak argument.

Kevin Lowe is actually a perfect example of a good team defensive defenseman who did get AST credit for what he was doing. I mean...Lowe was a fine defenseman, but I would be hard-pressed to call him better than the voting numbers indicate. The best players at all positions tend to be effective going both ways. There is no doubt that, since expansion, pure offensive defensemen have fared better in AST voting than pure defensive defensemen - this is a clear distortion, and I do not believe, for example, that Phil Housley was really a better hockey player than Kevin Lowe.

But I don't believe, at all, that defensive defensemen have been underrated in comparison to guys who were two-way players. When this line of thinking gets to the point of aggrandizing a guy like McCrimmon in comparison to a guy like Beck (who was good for maybe 10 points/season more than McCrimmon, on average), I think it has outlived its usefulness.
I think it's very relevant, at least to recent years. I'm fairly certain than Konstantinov, Hatcher, and Stevens are the only defensemen who have been a post-season All Star while scoring fewer than 40 points since Rod Langway.

And it took Hatcher being 5th in plus/minus to get his 2nd Team All Star nod.

TheDevilMadeMe is online now  
Old
02-25-2012, 01:31 PM
  #272
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,512
vCash: 500
Quote:
Originally Posted by TheDevilMadeMe View Post
Is there a GM here who doesn't realize that a 17th place finish is going to only be a few votes?
I think it's important to realize that a 17th place finish in voting for a defenceman is in no way a consensus that the player was the 17th best. It's a byproduct of a system designed to select the best player (Norris) or the top 4 players (All-star) and it may be based on the opinion of a very small minority.

overpass is offline  
Old
02-25-2012, 01:33 PM
  #273
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 38,096
vCash: 500
Quote:
Originally Posted by overpass View Post
I think it's important to realize that a 17th place finish in voting for a defenceman is in no way a consensus that the player was the 17th best. It's a byproduct of a system designed to select the best player (Norris) or the top 4 players (All-star) and it may be based on the opinion of a very small minority.
I agree. But I think a player consistently getting a handful of votes is meaningful.

TheDevilMadeMe is online now  
Old
02-25-2012, 01:40 PM
  #274
Hawkey Town 18
Moderator
 
Hawkey Town 18's Avatar
 
Join Date: Jun 2009
Location: Chicago, IL
Posts: 4,159
vCash: 500
Quote:
Originally Posted by Nalyd Psycho View Post
To expand on this, way back when I was looking at selke votes for RWs, I found only three names got noteworthy support. Kurri, Lehtinen and Gustafsson. Lehtinen of course played both wings and Gustafsson seemingly played all three forward positions. But those three were the only players to take regular shifts at RW and get regular support from Selke voters.
Not that he was available, but what about Dirk Graham?

Hawkey Town 18 is offline  
Old
02-25-2012, 01:45 PM
  #275
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 2,982
vCash: 500
Quote:
Originally Posted by TheDevilMadeMe View Post
Is there a GM here who doesn't realize that a 17th place finish is going to only be a few votes?
Probably but it might be hard to spot when it's lumped in there to consiously trying to skew numbers to win an argument.

Hobnobs is offline  
Closed Thread

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 03:44 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.