Thread: Prospect Info: Timmins Hits 25
View Single Post
Old
04-05-2012, 07:23 PM
  #46
Mathletic
Registered User
 
Mathletic's Avatar
 
Join Date: Feb 2002
Location: St-Augustin, Québec
Country: Canada
Posts: 12,351
vCash: 500
Quote:
Originally Posted by LyricalLyricist View Post
That could work but how do we evaluate it in cases where it's close? I mean, what is the performance index here? Either way, it's such a hard metric to have because if you go by GP, not necessarily quality player. If you go by points, intangibles are lost. I suppose if you rate a player's overall ability on some kind of scale then multiply it by GP and a sort of draft location scale. For instance, 1st overall should have less impact on the quality of an organization's draft over 2nd overall and so on.

There's plenty of ways to do it, but I find in order to have an accurate assessment a LOT of trail and error on the draft class scale and draft position scale would have to be modified.

This is off the top of my head, but i'm sure you understand what I mean. The values would have to be tweaked and even then, they may not be foolproof but you'd hope if you have a large enough sample space, it would be reflective on the quality of a team's drafting.
When it's close simply mark it as a tie. You can set a threshold of +/- 20 or 50 games, it will basically come down to the same thing. You can use all sorts of metric. Win shares, adjusted +/- or whatever else, GP is still a good proxy. Not saying it's perfect to the point that the guy who played 501 games is absolutely better than the 500 games guy. But it will get it done on average.

Mathletic is offline   Reply With Quote