View Single Post
Old 04-07-2018, 06:45 AM   #486
DeluxeMoustache
 
DeluxeMoustache's Avatar
 
Join Date: Mar 2010
Exp:
Default

Quote:
Originally Posted by Itse View Post
It's both. Goal differential and gf% are excellent predictors of future success.
Maybe think of it as that the objective is to score goals for and prevent goals against
- the idea is to get shots on net, usually required to score
- goalie should stop 90 percent of shots so keep shooting
- scoring is a result of shooting

The idea behind Corsi is quantifying the proportion of behaviour (shots) that lead to the expected result (goals).

So some people are saying GF is a result based stat. Fair, especially relative to Corsi

You are saying it is a predictor of standings results.
They are trying to add a previous level which is arguably more behavioural in terms of fundamentals and less outcome based

Or in other words, consider it in terms of dependence stepwise
- more shots proportionally presumably should result in more goals proportionally, all else being equal
- more goals proportionally should result in more wins, all else being equal
- more wins should result in better place in the standings, all else being equal

So of course goal differential (step 2) should be a better predictor of place in standings (step 3) than shot differential (step 1), because the ‘all else being equal’ qualifier drops off as you move up each step.

Last edited by DeluxeMoustache; 04-07-2018 at 06:52 AM.
DeluxeMoustache is offline   Reply With Quote
The Following User Says Thank You to DeluxeMoustache For This Useful Post: