View Single Post
Old
06-28-2012, 01:06 PM
  #38
jfb392
Registered User
 
Join Date: Jul 2010
Country: Canada
Posts: 8,121
vCash: 500
Quote:
Originally Posted by FreddieBisco View Post
At some point these micro analyzing stats need to stop being used over real production , goals, assists, points.

These are all fine and dandy but who is to say that when Ennis gets tougher minutes next season that he won't produce more? Is there a stat that predicts that? I'm sorry, Ennis scored 34 points in 48 games. He was the former ahl Roty. He led our team in playoff scoring against Boston. I don't think Ennis' accomplishments should be discredited due to stats that are controlled by some web site that is not affiliated with the NHL.

Using this logic, what's better a guy scoring 70 points playing the 12th "hardest" minutes or a guy who scores 40 points but against the "toughest" competition? Who is to say the guy who scores 40 points playing the toughest competition would be able to score 70 then playing the weakest? There is no way to tell and that's why these subjective stats don't tell you more then the real stats, goals + assists.
The majority of advanced statistics are simply re-purposed data from NHL.com, so it's not like they are privately recorded or something.
The data is just scraped and formulas are applied.

I fear that this concept may be difficult for you to grasp, however.
It's much easier to completely ignore them.

Quote:
Originally Posted by Der Jaeger View Post
Admittedly, the folks on this board have taught me a lot about advanced stats. It's new to me as of about two months ago.

With that said, I agree with all of Coller's analysis except for the puck possession.

Corsi is the one stat that I perceive to be misleading, somewhat. It does measure shots for and against while a player is on the ice. That's not a direct translation of puck possession, nor does it measure scoring chances.

When I watch Detroit play, I notice that many of Detroit's scoring chances are very good, due to the skill of their players. There defense is positioned well, mostly, so they may give up shots, but those shots might not be high quality.

In the same way, Ennis' skill and speed put a lot of pressure on opposing defenses, and created a lot of high quality scoring chances. And how many times did we see Ennis wheeling in the offensive zone with opposition unable to get the puck away from him?

I think the corsi = puck possession isn't totally valid, though it's about as close as we can get. And with this point in mind, I don't think that the opposition was able to possess the puck over Buffalo in the way Coller writes about it.
Corsi is actually a fine predictor of both puck possession and scoring chances.
In fact, recent research has shown that it and Fenwick are essentially as accurate as actual recorded scoring chances over larger sample sizes:
Quote:
Whatever tendency certain players might have for driving their team to get more scoring chances than a simple shot differential predicts is small and swamped by random noise. This suggests tracking scoring chances isn't adding much information to the readily available shot differential numbers.
http://nhlnumbers.com/2012/6/26/shot...nd-shot-totals

jfb392 is online now   Reply With Quote