View Single Post
Old
01-22-2013, 10:16 PM
  #417
Malkin4Top6Wingerz
Can you like, shutup
 
Malkin4Top6Wingerz's Avatar
 
Join Date: Mar 2009
Country: United States
Posts: 5,029
vCash: 500
Quote:
Originally Posted by JTG View Post
You mean defending Fleury on the basis that he has shown the ability to backstop a team to multiple Cup finals, and win one? Yeah sure...
I'll take that as a yes, and then respond to your previous post with that in mind.

Quote:
Originally Posted by JTG View Post
Hockey isn't baseball. You can't individualize a players stats in hockey, aside from goals and assists, and even then, those stats can be somewhat misleading. It's the ultimate team game, and trying to use an advanced statistic isn't going to work.

James Neal taking a shot shouldn't have an outcome on Paul Martin's advanced statistic. Paul Martin blowing coverage and allowing a guy to score a goal, shouldn't **** up James Neal's +/-. That's the problem I have with hockey statistics.
What is your position on QB rating in the NFL? It's an individual statistic but is also largely team influenced. A reciever can drop a wide open pass, or tip it directly to the opposing team and cost his QB an interception. Offensive lineman can blow their assignments and force their QB to make poor decisions. Yet, by and large, QB rating is an excellent indicator of a QB's effectiveness and ability level. Only the best QBs can maintain QB high ratings from season to season. The underlying point being that an individual stat that is directly affected by the players around them can still hold a lot of value.

Back to baseball, ERA is a stat that is largely driven by individuals but is still subject to a team's fielding ability. A team with a couple of gold glovers will be able to get to balls that other fielders wouldn't. That makes ERA an imperfect measure of ability. However imperfect != not valuable. If you were to have a bunch of pitchers with low career ERAs you would undoubtedly have an elite pitching staff.

James Neal taking a shot with Paul Martin on the ice isn't a random event. Paul Martin likely influenced this with his ability to defend and get the puck up the ice. If he's only maintaining a strong Corsi on account of his talented teammates, there are other advanced numbers which measure quality of linemates that we can use. We can also look at his Corsi relative to his own teammates, and WOWY (with you without you) numbers tell us what a player's Corsi is with and without specific players on the ice. All of these statistics can be used to get a solid idea of how valuable a player is. Is it perfect? No, but nothing is, and it's more valuable than the eyetest of 99% of HFers. Corsi is also not subject to noise the way plus minus is because of how many statistical events it measures. If James Neal gets off a shot that had nothing at all to do with anything Paul Martin did, it doesn't matter because it's only one of thousands of shot attempts during a season. It's not going to make an impact when all is said and done.

Quote:
Originally Posted by Rowdy Roddy Peeper View Post
Lovejoy and Martin's relative quality isn't the issue though. I only used Martin as an example to show the folly of slavishly following advanced statistics to determine quality.

Martin had the best Corsi of all our top 4 defensemen (including Letang) but he was atrocious last year by everyone's admission.
Corsi isn't a perfect measure of ability or how well a player is playing. But a consistently strong Corsi is the sign of a player who can influence goal differential at even strength. Paul Martin is very good at that and always has been, even during a down year. If you look at it through that lens instead of a tell all statistic you'll see that it has plenty of uses.

Malkin4Top6Wingerz is offline   Reply With Quote