HFBoards

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

Chris Kelly appreciation thread

Reply
 
Thread Tools
Old
03-21-2007, 08:27 PM
  #51
Herdin
Registered User
 
Join Date: Oct 2005
Location: Ottawa
Country: Canada
Posts: 217
vCash: 500
Quote:
Originally Posted by Lexicon Devil View Post
I wouldn't say the method is deeply flawed. There are some problems. The first is sample size - Heatley and Schubert have just over 60 minutes on the PK, so I wouldn't put too much into their stats.

But the fact that Redden plays a regular PK shift and has only been on the ice for 7 PP goals against - that has to mean something.

The other potential problem is quality of competition. This is tracked on the site, and it shows that all the defencemen face roughly the same quality of competition, but then again so does the entire league. The problem with this measure is that if 2nd unit PK guys on every team faceoff against 2nd unit PP guys, the competitive balance will be roughly the same as the 1st unit PK guys against the 1st unit PP guys. As a result, everyone will have roughly the same +/- stats. Then it is impossible to distinguish between quality of competition based on +/-.
Good point. Another problem to consider is that 5 on 3 PKs typically are far more difficult and usually your top PK guys will be facing it, which results in more minuses for them.

I'll retract the bit about it being deeply flawed but I'm still unconvinced that we can really draw anything useful from it without making far more specific examinations such as finding out how difficult of an opposition they faced for certain minutes, what type of PK situation they were in (5on4 vs 50n3), and how well the goalie was playing during that time (although that can become a chicken and egg situation).

Herdin is offline   Reply With Quote
Old
03-22-2007, 10:25 AM
  #52
gabe
 
Join Date: Feb 2007
Posts: 27
vCash: 500
Quote:
Originally Posted by Herdin View Post
Good point. Another problem to consider is that 5 on 3 PKs typically are far more difficult and usually your top PK guys will be facing it, which results in more minuses for them.

I'll retract the bit about it being deeply flawed but I'm still unconvinced that we can really draw anything useful from it without making far more specific examinations such as finding out how difficult of an opposition they faced for certain minutes, what type of PK situation they were in (5on4 vs 50n3), and how well the goalie was playing during that time (although that can become a chicken and egg situation).
The numbers on the site are all 5-on-4.

What do you think would be a good way to measure difficulty of opposition?

gabe is offline   Reply With Quote
Old
03-22-2007, 01:37 PM
  #53
Lexicon Devil
Registered User
 
Lexicon Devil's Avatar
 
Join Date: Apr 2002
Posts: 8,343
vCash: 500
Quote:
Originally Posted by gabe View Post
What do you think would be a good way to measure difficulty of opposition?
For the PK stats it might be interesting to look specifically at the average PP icetime of the opponents. The idea is to go by the opposing coach's idea of who their best PP guys are. The better guys get around 4.5-5 mins/game, while the second unit guys on most teams are around 2.5-3.5 mins/game.

So if we see that Kelly's opponents average 4.15 mins/game while Heatley's average 2.75 mins/game, this would seem quite telling.

You might be able to do something similar for the PP, but I suspect that icetime is more evenly distributed amongst PKers (at least forwards).


I realize this is a completely different way of measuring difficulty of competition than what you use for even-strength (which seems to work reasonably well). However you could always use both, and crosscheck to see how well they correlate.


Last edited by Lexicon Devil: 03-22-2007 at 01:43 PM.
Lexicon Devil is offline   Reply With Quote
Old
03-22-2007, 04:57 PM
  #54
gabe
 
Join Date: Feb 2007
Posts: 27
vCash: 500
Quote:
Originally Posted by Lexicon Devil View Post
For the PK stats it might be interesting to look specifically at the average PP icetime of the opponents. The idea is to go by the opposing coach's idea of who their best PP guys are. The better guys get around 4.5-5 mins/game, while the second unit guys on most teams are around 2.5-3.5 mins/game.

So if we see that Kelly's opponents average 4.15 mins/game while Heatley's average 2.75 mins/game, this would seem quite telling.

You might be able to do something similar for the PP, but I suspect that icetime is more evenly distributed amongst PKers (at least forwards).


I realize this is a completely different way of measuring difficulty of competition than what you use for even-strength (which seems to work reasonably well). However you could always use both, and crosscheck to see how well they correlate.
That's a good idea. I'll try it out.

The correlation between minutes and PP+/- is 0.35, so the results may not come out substantially different than before, but at least everyone in the league will be on approximately the same scale.

gabe 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 09:01 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.