Quote:
Originally Posted by Enoch Root
If you want to try and quantify draft success via games played per pick, the picks have to be factored by what round they occurred in. You can't compare a top 5 pick (expectation of what 500 - 1000 games?) vs a 4th rounder (expectation of 200 games X 10% chance of hitting = 20 games).
You need a scale for the draft spot, to pro-rate the expectation for that pick, then calculate games per pro-rated picks
(You could use the chart that compares the value of picks, for instance)
|
Case in point, Yakupov had 350 NHL games which would be pretty successful for a later round pick, but a failure for a 1st rounder. Same thing with Puljuarvi. Jankowski is another one. When all is said and done, he will have had a decent career as a role player, but no one would call that a great pick overall.
And sometimes bad teams have to play the players they drafted even if they wouldn't make most NHL rosters. Being the best of a bad bunch, doesn't make them good picks, then try to make an objective comparison.
I don't think there is a simple way to model draft success. I think the best you can do is compile samples where there are clear hits and clear failures.