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

A new look at plus/minus

Reply
 
Thread Tools
Old
01-31-2011, 07:41 PM
  #1
HCH
Registered User
 
Join Date: Dec 2003
Location: The Wild West
Country: Canada
Posts: 4,655
vCash: 500
A new look at plus/minus

The plus/minus stat is one of the most misleading stats used in the NHL today. It does not account for the unequal distribution of ice time. Nor does it take into account that every time a goal is scored at even strength, all five play players get a minus even though only one goal is scored. Finally, it does not take into account the relative strength of the team.

A player can be reasonably good but if he is on an awful team, he gets painted with a black brush. In other words, does the player suffer from playing on a bad team.

I tried to come up with a formula that took this into account and there were some interesting results. The regular plus minus stats are in brackets and the adjusted rating follows. The rating is not a goal differential it is only intended for comparative purposes.

The first step is to equalize even strength ice time and then attribute only 1/5 of a goal against to each player. The second step is to adjust the +/- for each forward to 15 minutes per game played and 20 minutes for each defenseman (4 lines, 3 defense, 60 minutes). Finally a comparison is made between the overall team +/- to the individual plus minus to minimize the impact that the rest of the team has on the individual +/-

Obviously it does not take into account the special skills some players have on the PP. It is only intended for even strength comparison and while it is far from perfect, I believe it gives a far superior reading on effectiveness than than +/-. Here is a short sample.

If someone would like to suggest a better method or some refinements, I would be happy to hear them.

Roman Hamrlik (2) rating = 4.6
Hal Gill (-4) rating = -8.7

Tomas Plekanec (10) rating = 34.5
Scott Gomez (-9) rating = -18.6

And here's the kicker. James Wiesnewski's Islander stats (doesn't include Habs stats). It reveals that the Wiz was actually less of a liability than a regular +/- would indicate.

James Wiesnewski (-18) rating = 9.8

HCH is offline   Reply With Quote
Old
01-31-2011, 08:41 PM
  #2
ScottFC
Registered User
 
ScottFC's Avatar
 
Join Date: Feb 2008
Location: чуд
Country: Canada
Posts: 1,107
vCash: 500
There is almost the exact same formula in a book called the economics of hockey, check it out.

ScottFC is offline   Reply With Quote
Old
01-31-2011, 10:08 PM
  #3
Dr. Charles
Registered User
 
Dr. Charles's Avatar
 
Join Date: Jan 2005
Location: Cracktown
Country: Canada
Posts: 733
vCash: 500
Quote:
Originally Posted by HCH View Post
The plus/minus stat is one of the most misleading stats used in the NHL today. It does not account for the unequal distribution of ice time. Nor does it take into account that every time a goal is scored at even strength, all five play players get a minus even though only one goal is scored. Finally, it does not take into account the relative strength of the team.

A player can be reasonably good but if he is on an awful team, he gets painted with a black brush. In other words, does the player suffer from playing on a bad team.

I tried to come up with a formula that took this into account and there were some interesting results. The regular plus minus stats are in brackets and the adjusted rating follows. The rating is not a goal differential it is only intended for comparative purposes.

The first step is to equalize even strength ice time and then attribute only 1/5 of a goal against to each player. The second step is to adjust the +/- for each forward to 15 minutes per game played and 20 minutes for each defenseman (4 lines, 3 defense, 60 minutes). Finally a comparison is made between the overall team +/- to the individual plus minus to minimize the impact that the rest of the team has on the individual +/-

Obviously it does not take into account the special skills some players have on the PP. It is only intended for even strength comparison and while it is far from perfect, I believe it gives a far superior reading on effectiveness than than +/-. Here is a short sample.

If someone would like to suggest a better method or some refinements, I would be happy to hear them.

Roman Hamrlik (2) rating = 4.6
Hal Gill (-4) rating = -8.7

Tomas Plekanec (10) rating = 34.5
Scott Gomez (-9) rating = -18.6

And here's the kicker. James Wiesnewski's Islander stats (doesn't include Habs stats). It reveals that the Wiz was actually less of a liability than a regular +/- would indicate.

James Wiesnewski (-18) rating = 9.8
I will have a much better sleep at night now that I know some people working hard on re-thinking existing stats.

Dr. Charles is offline   Reply With Quote
Old
01-31-2011, 11:02 PM
  #4
llamateizer
Registered User
 
llamateizer's Avatar
 
Join Date: Mar 2007
Location: Montreal
Country:
Posts: 5,426
vCash: 500
i dont understand.
if you get a -5 playing 20 min/game
vs -5 playing 10 min/game... what does it change?

how scott gomez gets a -18?
what -18 represent?

wiz is better than Hamrlik?

llamateizer is offline   Reply With Quote
Old
01-31-2011, 11:15 PM
  #5
Kriss E
Registered User
 
Join Date: May 2007
Posts: 24,038
vCash: 500
Really makes zero sense to me...

Kriss E is offline   Reply With Quote
Old
01-31-2011, 11:20 PM
  #6
MasterDecoy
Carlos Danger
 
MasterDecoy's Avatar
 
Join Date: May 2010
Location: Beijing
Posts: 10,099
vCash: 500
do picard!

MasterDecoy is offline   Reply With Quote
Old
02-01-2011, 06:34 AM
  #7
swissexpert
Registered User
 
swissexpert's Avatar
 
Join Date: Sep 2009
Country: Switzerland
Posts: 1,159
vCash: 500
Quote:
Originally Posted by MasterDecoy View Post
do picard!
Yeah, please give a comparison between Picard/Weber or do all defensemen. Only the fact that Wiz is a + player with your method doesn't mean anything...

swissexpert is offline   Reply With Quote
Old
02-01-2011, 07:58 AM
  #8
onice
Registered User
 
Join Date: Jan 2004
Location: Montreal
Posts: 5,514
vCash: 500
HCH,

Can you show us the math you went through for just one player. Gomez let's say.

Although I noticed that your system proves my point: Gomez is worse than useless, I still would like to see how you quantified it. I get a little suspicious when people throw stats or figures at me and don't tell me how they arrived at those figures.

Some parts of your premise are redundant. For instance whether you assign a -1 or a -1/5 to a goal against doesn't really matter as long as you are consistent. What intrigues me is how you converted the +/- according to TOI. I'd like to see the math in that. Also I'd like to see how you related an individual's total to the +/- of the whole team. Show us the math you used. Don't do it for all the players, just do it for one player.

onice is offline   Reply With Quote
Old
02-01-2011, 09:26 AM
  #9
Dr. Charles
Registered User
 
Dr. Charles's Avatar
 
Join Date: Jan 2005
Location: Cracktown
Country: Canada
Posts: 733
vCash: 500
Let's re-think all the stats : Every player should get a point when a goal is scored because it's a team effort.

Dr. Charles 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 02:36 PM.

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.