Jump to content

Photo

Yet Another Reason Why Plus/Minus Is Stupid


  • Please log in to reply
27 replies to this topic

#21 ohms law

ohms law

    Senior Devil

  • Members
  • PipPipPipPip
  • 646 posts

Posted 04 January 2012 - 12:38 PM

You know, +/- isn't a completely terrible stat for defensemen. For forwards is pretty much a complete crap shoot, but we all know that defensemen generate different kinds of statistics.
humm
  • 0

#22 Triumph

Triumph

    A Legend

  • Mod
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 26,893 posts

Posted 04 January 2012 - 01:09 PM

You know, +/- isn't a completely terrible stat for defensemen. For forwards is pretty much a complete crap shoot, but we all know that defensemen generate different kinds of statistics.
humm


What does this even mean? Anyway: http://vhockey.blogs...st-v-trees.html

+/- is bad for defensemen for the same reason it's bad for forwards - it takes into account absolutely 0 context.
  • 0

http://drivingplay.blogspot.com - The blog with three first lines
 


#23 EdgeControl

EdgeControl

    All-Star

  • Members
  • PipPipPipPipPip
  • 1,280 posts

Posted 04 January 2012 - 01:18 PM

kovalchuk got a +1 for his penalty shot....

who got the minus'????
  • 0

#24 ohms law

ohms law

    Senior Devil

  • Members
  • PipPipPipPip
  • 646 posts

Posted 04 January 2012 - 01:23 PM

What does this even mean? Anyway: http://vhockey.blogs...st-v-trees.html

+/- is bad for defensemen for the same reason it's bad for forwards - it takes into account absolutely 0 context.


Good link. Thanks.

...you know, humm. I'm still reading, but as soon as I see things like: "By way of example, in 08/09 Chris Pronger had a 5v5 on-ice save percentage of .915. When he was in the game, but not on the ice, the opponents scored at a 2.20 goals per 60 clip..." it makes me start to ask questions. This is amateurish statistical thinking. But, I'll give it a chance. Be back later.
  • 0

#25 ohms law

ohms law

    Senior Devil

  • Members
  • PipPipPipPip
  • 646 posts

Posted 04 January 2012 - 01:42 PM

I don't get it. This person (the author) is saying things like:
"So the correlation of 08/09 Corsi QualComp to 09/10 5v5 save percentage score is a touch stronger, r=-.09."

and we're supposed to listen to him? Has he ever taken a statistics class, even? :blink:
  • 0

#26 Devils731

Devils731

    A Legend

  • Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 17,436 posts

Posted 04 January 2012 - 05:00 PM

I don't get it. This person (the author) is saying things like:
"So the correlation of 08/09 Corsi QualComp to 09/10 5v5 save percentage score is a touch stronger, r=-.09."

and we're supposed to listen to him? Has he ever taken a statistics class, even? :blink:


I'm not sure I understand you're problem with his statement. He's saying that there isn't much correlation to begin with, but that the negative correlation got a little stronger when comparing 2 seasons, as opposed to just one.

Guys that face tough competition and do well tend to face tougher competition the next season, which hurts their on ice save percentage, but it's such a negligible effect that it should effectively be ignored, is how I read it.
  • 0
Your unconditional rejection of violence makes you smugly think of yourselves as noble, as enlightened, but in reality it is nothing less than abject moral capitulation to evil. Unconditional rejection of self-defense, because you think its a supposed surrender to violence, leaves you no resort but begging for mercy or offering appeasement.

-Terry Goodkind


Sex Panther cologne -- 50 percent of the time, it works every time.

-Anchorman

The best time to plant a tree is 20 years ago. The second best time is now.

-Anonymous

Keeper of Section 212-213's wayward step

#27 ohms law

ohms law

    Senior Devil

  • Members
  • PipPipPipPip
  • 646 posts

Posted 04 January 2012 - 07:01 PM

(I'm getting crap for being passive-agressive in the other thread, so here's a non-PA reply:)

Dude, a +/- 9% "correlation" is not a correlation at all. Stating that one correlation is better than another, when both are under 80% to begin with, is just ignorant.
I got a laugh over the author's scatter plot analysis as well. If this is the kind of analysis that is being listened to around here... well, there goes a bunch of respect out of the window. That article was just stupid.
  • 0

#28 Triumph

Triumph

    A Legend

  • Mod
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 26,893 posts

Posted 04 January 2012 - 07:22 PM

(I'm getting crap for being passive-agressive in the other thread, so here's a non-PA reply:)

Dude, a +/- 9% "correlation" is not a correlation at all. Stating that one correlation is better than another, when both are under 80% to begin with, is just ignorant.
I got a laugh over the author's scatter plot analysis as well. If this is the kind of analysis that is being listened to around here... well, there goes a bunch of respect out of the window. That article was just stupid.


I can assure you the author knows exactly what he's talking about. Feel free to write a competing article about whether defensemen can affect save percentage on - I'd actually be really interested to read that, it would change the way I and a lot of other people are thinking about hockey these days.
  • 0

http://drivingplay.blogspot.com - The blog with three first lines
 





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users