View Single Post
05-02-2013, 03:21 PM
Registered User
SnowblindNYR's Avatar
Join Date: Nov 2011
Location: Philadelphia, PA
Posts: 25,281
vCash: 500
Originally Posted by Blue Blooded View Post
HART is definitely a stat that cannot be looked upon and said: this guy is better than that guy because his rating was .001 better! Too many variables. It's hard to accurately project the impact of the quality of the players on the ice on a players' stats. However, not doing it could perhaps distort the picture even more.

I usually only try to compare players that have been used similarly when using adv. stats, unless there is too great a difference in some key stats where one could say that player A has been superior to player B.

But a good rating is definitely not a negative in any case!

Like Snowblind's example of GF%; if Stepan is #1 and Ribeiro is #200 it is hard to dispute that Stepan has been a superior ES player.

HART among skaters with 500+ minutes: Stepan #3, Ribeiro #204, Bäckström #115, Richards #50, Brassard #78.
You know it's nice to look at, and it's also nice to be able to brag that our 3rd line center had a significantly higher HART than their super ultra elite #1 center, however as you said "too many variables". Too many variables means you get things like a solid but unspectacular defenseman on the best team in the league is 3rd. Just too much chance for error. Too many possible loose correlations. Too many factors that affect each of the variables. That's why I like GF% because it reduces the variables and thus chances for weird results.

SnowblindNYR is offline