HFBoards

Go Back   HFBoards > Fantasy Hockey Talk > All Time Draft
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
All Time Draft Fantasy league where players of the past and present meet.

ATD 2013 - Draft Thread III

Closed Thread
 
Thread Tools
Old
02-07-2013, 01:08 AM
  #151
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Babe Dye is one of the players who I am planning to profile during the course of the draft. He's another pre-war star about whom I think not enough is known. I'll post what I've got when I have the time to pull it all together.

Sturminator is offline  
Old
02-07-2013, 01:25 AM
  #152
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Quote:
Originally Posted by Hawkey Town 18 View Post
Are these vs. #1? Or maybe they don't account for Lemieux?

The vs. #2 file I have has: 100, 94, 94, 91, 78, 70, 69, 67

Either way, I don't think he's that far behind Schriner, but he is definitely behind him.
Those are the numbers you get by including Lemiuex/Gretzky and only dropping from #2 to #3 if there is greater than a 10% difference between #2 and #3 scoring. This is my preferred method, and I think it is more legitimate than simply dropping players based on their name.

The point of the system is to remove outliers, not to remove people. The outlier here is not the mere existence of Gretzky and Lemieux, but rather their statistical dominance in their best seasons. But they were not statistical outliers in every season, so why should they be consistently removed? I see how you got your numbers, but I think there is a better method.

1996-97:

Lemieux: 122
Selanne: 119
Kariya: 109

You have this season as a 100 Vs2 for Kariya because Mario is the name at the top. But Mario is already removed by the system because he was the #1 scorer in the league. You essentially remove the outlier twice - treating Mario as if he never existed, and then Selanne as the #1 scorer, who is then removed, himself. I don't think this is a very logical method.

Actually, the above is sort of hard luck for Kariya under the better method (10% rule) because Selanne is almost the full 10% above Kariya, so Paul ends up having almost the worst possible Vs2 score as a #3 scorer in the league: 92.

I think it may be time to revive the Vs2 thread so we can have an organized discussion with these numbers getting thrown around more often now.

---------------------------------------------------------------------

Regarding Kariya vs. Schriner, the argument in favor of Paul is that he was less of a floater, and has a quite strong Hart record. I like Schriner a bit better, but I think they're close.


Last edited by Sturminator: 02-07-2013 at 02:19 AM.
Sturminator is offline  
Old
02-07-2013, 04:15 AM
  #153
DoMakc
Registered User
 
Join Date: Jun 2006
Posts: 938
vCash: 500
Quote:
Originally Posted by Sturminator View Post
Those are the numbers you get by including Lemiuex/Gretzky and only dropping from #2 to #3 if there is greater than a 10% difference between #2 and #3 scoring. This is my preferred method, and I think it is more legitimate than simply dropping players based on their name.

The point of the system is to remove outliers, not to remove people. The outlier here is not the mere existence of Gretzky and Lemieux, but rather their statistical dominance in their best seasons. But they were not statistical outliers in every season, so why should they be consistently removed? I see how you got your numbers, but I think there is a better method.

1996-97:

Lemieux: 122
Selanne: 119
Kariya: 109

You have this season as a 100 Vs2 for Kariya because Mario is the name at the top. But Mario is already removed by the system because he was the #1 scorer in the league. You essentially remove the outlier twice - treating Mario as if he never existed, and then Selanne as the #1 scorer, who is then removed, himself. I don't think this is a very logical method.

Actually, the above is sort of hard luck for Kariya under the better method (10% rule) because Selanne is almost the full 10% above Kariya, so Paul ends up having almost the worst possible Vs2 score as a #3 scorer in the league: 92.

I think it may be time to revive the Vs2 thread so we can have an organized discussion with these numbers getting thrown around more often now.

---------------------------------------------------------------------

Regarding Kariya vs. Schriner, the argument in favor of Paul is that he was less of a floater, and has a quite strong Hart record. I like Schriner a bit better, but I think they're close.
I still don't like vs2 method because it's so arbitrary (as you can see at the whole on the whole exclution debate: Gretzky, Lemieux, their teammates, their neighbors, their pets). i like percentile approach more, but then , it's still arbitrary: which one should choose - 5%, 10%, but in the you adjust your percentile according to your opinion on given player. I method I like most is baseball OPS+, where you don't adjust for greatness to measure mediocrity, but rather adjust for mediocrity of the era, so you can measure greatness.

DoMakc is offline  
Old
02-07-2013, 04:35 AM
  #154
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Quote:
Originally Posted by DoMakc View Post
I still don't like vs2 method because it's so arbitrary (as you can see at the whole on the whole exclution debate: Gretzky, Lemieux, their teammates, their neighbors, their pets). i like percentile approach more, but then , it's still arbitrary: which one should choose - 5%, 10%, but in the you adjust your percentile according to your opinion on given player. I method I like most is baseball OPS+, where you don't adjust for greatness to measure mediocrity, but rather adjust for mediocrity of the era, so you can measure greatness.
No method is perfect. I think the percentile method is better because it is not arbitrary, but it has its flaws, as well. If you have a suggestion for a better method (I have no idea how OPS+ works), please share it.

I think the problem with the exclusion debate is that some people have just been a bit mixed up about the whole point of the Vs2 system, which is to remove outliers and compare player scoring against something resembling a common baseline. Certain players are outliers on a career level, but that does not mean they should be arbitrarily removed on a year-to-year basis simply because they had great careers, especially when they are already the #1 scorer, and thus accounted for the by existing Vs2 methodology (as in the above example with Mario).

I think almost anyone, if they really think about it, would agree that a percentile exclusion system is better than picking and choosing names to leave out of the analysis.

edit: just what the percentile threshhold should be and just how we should handle big dropoffs from #2 to #3 is obviously still open for debate.

Sturminator is offline  
Old
02-07-2013, 04:45 AM
  #155
DoMakc
Registered User
 
Join Date: Jun 2006
Posts: 938
vCash: 500
Quote:
Originally Posted by Sturminator View Post
No method is perfect. I think the percentile method is better because it is not arbitrary, but it has its flaws, as well. If you have a suggestion for a better method (I have no idea how OPS+ works), please share it.

I think the problem with the exclusion debate is that some people have just been a bit mixed up about the whole point of the Vs2 system, which is to remove outliers and compare player scoring against something resembling a common baseline. Certain players are outliers on a career level, but that does not mean they should be arbitrarily removed on a year-to-year basis simply because they had great careers, especially when they are already the #1 scorer, and thus accounted for the by existing Vs2 methodology (as in the above example with Mario).

I think almost anyone, if they really think about it, would agree that a percentile exclusion system is better than picking and choosing names to leave out of the analysis.

edit: just what the percentile threshhold should be and just how we should handle big dropoffs from #2 to #3 is obviously still open for debate.
OPS+ uses league average as a benchmark, like vs2 uses the second place finish. Basically shows how many per cent a player's OPS (points for hockey) was better than an average player's in every given year.
OPS+ also adjusts for park factor, but I think only Gilbert Perreault fans would like to do it for hockey.

DoMakc is offline  
Old
02-07-2013, 05:17 AM
  #156
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Quote:
Originally Posted by DoMakc View Post
OPS+ uses league average as a benchmark, like vs2 uses the second place finish. Basically shows how many per cent a player's OPS (points for hockey) was better than an average player's in every given year.
OPS+ also adjusts for park factor, but I think only Gilbert Perreault fans would like to do it for hockey.
The problem with trying to use a baseball-style league average scoring metric in hockey is that production in baseball is discreetly parsed into at-bats (or for pitchers, batters faced), which can be used as the quanta of an averaging system. In hockey, this is impossible.

Average scoring among which players? Do we take icetime into account, and if so, how? What about special teams usage? Baseball is mostly a series of matchups between individuals, with very little actual team dynamic involved and no real "special" game situations. As such, it is relatively easy to seperate individual performance from statistical noise. Hockey is a deeply-integrated team sport, and thus full of noise. We are left having to use more primitive methods.

Sturminator is offline  
Old
02-07-2013, 06:42 AM
  #157
BubbaBoot
Registered User
 
BubbaBoot's Avatar
 
Join Date: Oct 2003
Location: The Fenway
Country: Wallis & Futuna
Posts: 11,066
vCash: 500
Quote:
Originally Posted by Sturminator View Post
The problem with trying to use a baseball-style league average scoring metric in hockey is that production in baseball is discreetly parsed into at-bats (or for pitchers, batters faced), which can be used as the quanta of an averaging system. In hockey, this is impossible.

Average scoring among which players? Do we take icetime into account, and if so, how? What about special teams usage? Baseball is mostly a series of matchups between individuals, with very little actual team dynamic involved and no real "special" game situations. As such, it is relatively easy to seperate individual performance from statistical noise. Hockey is a deeply-integrated team sport, and thus full of noise. We are left having to use more primitive methods.
Very nice argument as to why I have the difficulty whenever I peruse through the different eras. Yes baseball has it's eras with their inherent idiosyncrasies but is seemingly static compared to hockey. There are so many variables that can't as thoroughly defined as a gauging point in hockey.

BubbaBoot is offline  
Old
02-07-2013, 07:39 AM
  #158
VanIslander
Hope for better 2015
 
VanIslander's Avatar
 
Join Date: Sep 2004
Posts: 19,614
vCash: 500
Hershey drafts 6'1, 210 lbs. left winger Luc Robitaille, whose first seven seasons were phenomenal with seven consecutive 1st or 2nd team all-star selections and seven straight years top-10 in goals. By age 26 he had already scored 382 NHL goals, including 34 playoff goals. He scored 799 points in his first 640 NHL games, not once in his first eight seasons dipping below the 84 point total of his rookie year. "Lucky" Luc worked hard to put himself in good scoring position and had a fantastic shot. He was known for his competiveness, even in practices, as much as he was known for his friendliness. He had a boatload of determination and it showed. He went on to play for over a decade more, including three more times top-10 in goals, five more times top-10 in powerplay goals, another 2nd team all-star. He is presently still top-10 all-time in career goals scored among all NHLers, tied with Jaromir Jagr. The fact that he has scored the most goals and points in history for a left winger is almost beside the point, as is his secondary role in a Stanley Cup championship in his 16th year.



Quote:
Originally Posted by Los Angeles Kings Legends
Robitaille made up for any skating deficiencies with one of the most accurate shots in NHL history. He was a regular leader in shooting percentage, thanks to a number of reasons. He worked himself into high percentage scoring areas, often down low and in tight. Though a defender might have been draped all over him, he always kept his stick unchecked. He would release his shot in the blink of an eye, usually just burying passes and rebounds with no backswing at all.

Robitaille, an under-noticed physical player, continued to be almost unquestioningly the league's best left winger for 8 seasons, consistently scoring goals. He scored at least 44 goals in 8 consecutive seasons (only Gretzky and Mike Bossy had better streaks), and also managed to shake his playoff jinx as he became a genuine playoff threat in 1992 with 12 goals in 12 games and in 1993 when he was a major part of the Kings "Cinderella" Cup run.

Just one year after coming so close to winning Lord Stanley's Grail, the Kings missed the playoffs. Robitaille played for Canada's national team at the 1994 World Championship in Italy. It was Robitaille who scored the gold medal winning goal in a shootout, giving Canada its first world championship in 33 years.


Quote:
Originally Posted by Los Angeles Times, Nov. 10, 2009
... he personified everything good about this game and the undying power of hope backed by tireless effort.

"I've known Luc since he was 17, with the Hull Olympiques," Gretzky said as he traveled the red carpet before the [Hall of Fame induction] ceremony.

"I've said this before: with 'Rocket' Richard, Mike Bossy, Guy Lafleur and Mario Lemieux, there's nobody who wanted to score more desperately than Luc Robitaille. He made himself a Hall of Famer."


Quote:
Originally Posted by Scott Burnside, ESPN.com Nov 5th 2009
When Robitaille made the Kings out of training camp in 1986, legendary Kings center Marcel Dionne asked him what he wanted to do now that he was in Los Angeles. Did he want to see the sights, meet the stars? No, he wanted to play hockey, Robitaille told him. OK, then, Dionne responded, Robitaille could move in with him and his family.

"I never went anywhere," Robitaille said, recalling the steady routine of going from practice to Dionne's home to games and back again.

Blake likewise recalled a deeply competitive player beneath that happy-go-lucky exterior. He remembered how, even in practice, Robitaille had to score. Even if his turn during a drill appeared to be over, he would fish the puck out of the corner and still rip it into the net, often to the consternation of netminders such as Kelly Hrudey.

Pat Quinn was the coach in Los Angeles at the time, and he recalled looking beyond the skating to see "a great brain in there."

Pure speed? No. But Quinn said he still thinks Robitaille got from the corner to the net as quickly as any player, a testament to his anticipation and that big hockey brain.


Last edited by VanIslander: 02-07-2013 at 07:56 AM.
VanIslander is offline  
Old
02-07-2013, 08:14 AM
  #159
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 4,179
vCash: 500
Quote:
Originally Posted by VanIslander View Post
Hershey drafts 6'1, 210 lbs. left winger Luc Robitaille

Hobnobs is offline  
Old
02-07-2013, 08:44 AM
  #160
DoMakc
Registered User
 
Join Date: Jun 2006
Posts: 938
vCash: 500
Quote:
Originally Posted by Sturminator View Post
The problem with trying to use a baseball-style league average scoring metric in hockey is that production in baseball is discreetly parsed into at-bats (or for pitchers, batters faced), which can be used as the quanta of an averaging system. In hockey, this is impossible.

Average scoring among which players? Do we take icetime into account, and if so, how? What about special teams usage? Baseball is mostly a series of matchups between individuals, with very little actual team dynamic involved and no real "special" game situations. As such, it is relatively easy to seperate individual performance from statistical noise. Hockey is a deeply-integrated team sport, and thus full of noise. We are left having to use more primitive methods.
While it's true, I don't see how it's relevant for benchmark choice. You have the same problem comparing with nr. 2 - not the same amount of TOI or PPTOI. So i would take the average over all forwards, who played ceratin amount of games. Same with defencemen.

OPS is a per plate appearance measure, since we comparing point production we already using primitive method.

DoMakc is offline  
Old
02-07-2013, 09:50 AM
  #161
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 26,036
vCash: 500
But more ice time is typically earned by being a better player.

seventieslord is offline  
Old
02-07-2013, 09:54 AM
  #162
Velociraptor
Registered User
 
Velociraptor's Avatar
 
Join Date: May 2007
Location: Maritimes
Country: Canada
Posts: 10,772
vCash: 500
Trail selects, Allan Stanley, D



LF's bio from ATD 2012

Velociraptor is offline  
Old
02-07-2013, 09:55 AM
  #163
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 4,179
vCash: 500
Quote:
Originally Posted by seventieslord View Post
But more ice time is typically earned by being a better player.
Or being a more healthy player or being more usefull is certain situations or beign favoured by a biased coach or being a player on a big contract (the lalime - barrasso situation for example) or being on a stacked team that can afford to rest players.

Hobnobs is offline  
Old
02-07-2013, 10:03 AM
  #164
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
When comparing scoring among the league leaders, I see no reason to worry so much about icetime. And trying to average scoring among skaters who reach a certain games threshhold without any accounting for icetime or role is going to distort things terribly across eras. The number of skaters used regularly in games has steadily increassed through the years, most of the increase coming in low talent, low icetime grinders. That's going to blow up any attempt at averaging league scoring irrespective of role.

Sturminator is offline  
Old
02-07-2013, 10:05 AM
  #165
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 4,179
vCash: 500
Does anyone have a list from leaflander or will he be around?

Hobnobs is offline  
Old
02-07-2013, 10:09 AM
  #166
Velociraptor
Registered User
 
Velociraptor's Avatar
 
Join Date: May 2007
Location: Maritimes
Country: Canada
Posts: 10,772
vCash: 500
Quote:
Originally Posted by Hobnobs View Post
Does anyone have a list from leaflander or will he be around?
I just texted him, hopefully I'll hear back soon.

Velociraptor is offline  
Old
02-07-2013, 10:09 AM
  #167
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Leadlander selects Cy Denneneny.

Sturminator is offline  
Old
02-07-2013, 10:24 AM
  #168
BubbaBoot
Registered User
 
BubbaBoot's Avatar
 
Join Date: Oct 2003
Location: The Fenway
Country: Wallis & Futuna
Posts: 11,066
vCash: 500
Quote:
Originally Posted by velociraptor View Post
trail selects, allan stanley, d



lf's bio from atd 2012
bollocks!!!

BubbaBoot is offline  
Old
02-07-2013, 10:28 AM
  #169
BenchBrawl
joueur de hockey
 
BenchBrawl's Avatar
 
Join Date: Jul 2010
Posts: 10,207
vCash: 800
Quote:
Originally Posted by Sturminator View Post
Leadlander selects Cy Denneneny.
what the hell is this?

BenchBrawl is offline  
Old
02-07-2013, 10:29 AM
  #170
BenchBrawl
joueur de hockey
 
BenchBrawl's Avatar
 
Join Date: Jul 2010
Posts: 10,207
vCash: 800
Quote:
Originally Posted by VanIslander View Post
Hershey drafts 6'1, 210 lbs. left winger [COLOR="Sienna"]Luc Robitaille]
Luc Robitaille is definitely one of the nicest guys in interview in french.Very natural and friendly.

BenchBrawl is offline  
Old
02-07-2013, 10:45 AM
  #171
seventieslord
Moderator
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 26,036
vCash: 500
Quote:
Originally Posted by Sturminator View Post
When comparing scoring among the league leaders, I see no reason to worry so much about icetime. And trying to average scoring among skaters who reach a certain games threshhold without any accounting for icetime or role is going to distort things terribly across eras. The number of skaters used regularly in games has steadily increassed through the years, most of the increase coming in low talent, low icetime grinders. That's going to blow up any attempt at averaging league scoring irrespective of role.
I agree.

By the way, Leaf Lander texted me his pick and it was Brian Dennehy.

seventieslord is offline  
Old
02-07-2013, 11:10 AM
  #172
VanIslander
Hope for better 2015
 
VanIslander's Avatar
 
Join Date: Sep 2004
Posts: 19,614
vCash: 500
Quote:
Originally Posted by seventieslord View Post
By the way, Leaf Lander texted me his pick and it was Brian Dennehy.

VanIslander is offline  
Old
02-07-2013, 11:11 AM
  #173
VanIslander
Hope for better 2015
 
VanIslander's Avatar
 
Join Date: Sep 2004
Posts: 19,614
vCash: 500
196. Leaf Lander - Toronto Maple Leafs - re-pick needed

197. jkrx & KingForsberg - Philadelphia Firebirds - ON THE CLOCK

198. Bring Back Scuderi - Pittsburgh Athletic Club
199. VanIslander & Hedberg - Hershey Bears
200. DoMakc - HC Donbass
201. Murphy & gmm - Edmonton Mercurys
202. Hobnobs - Seattle Eskimos

VanIslander is offline  
Old
02-07-2013, 11:14 AM
  #174
Hobnobs
Pinko
 
Hobnobs's Avatar
 
Join Date: Nov 2011
Country: Sweden
Posts: 4,179
vCash: 500
I dont really see the point in vs2 either. It shows one thing and in most cases its a very minor stat to use to compare two players (or more)

Hobnobs is offline  
Old
02-07-2013, 11:25 AM
  #175
Sturminator
I voted for Kodos
 
Sturminator's Avatar
 
Join Date: Feb 2002
Location: West Egg, New York
Country: Ukraine
Posts: 8,037
vCash: 500
Quote:
Originally Posted by Hobnobs View Post
I dont really see the point in vs2 either. It shows one thing and in most cases its a very minor stat to use to compare two players (or more)
It is a very meaningful improvement on just using top-10 or top-20 or top-X scoring placements. It is not perfect, but it does a much better job of relativizing performance across eras than does any raw scoring or raw placement system.

Sturminator 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 10:59 PM.

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

vBulletin Copyright ©2000 - 2015, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2015 All Rights Reserved.