Ohashi_Jouzu*
Registered User
The reason I'm asking is because if Pnep's system is based on educated guesses/eyeballing the numbers (there's nothing wrong with that, of course), I'd like to try to develop a purely statistical regression model. I think I can do it, but I don't want to waste my time if Pnep has already beat me to it.
Would be tough to go pure statistical, I reckon. There are facts that have been brought up in other threads, such as no 2-time Hart winner has ever been denied. Seems like you have to value it pretty highly. Cup wins are also valued highly, but I would have to imagine that there are players with lots of Cups who might not get many points in other categories... and probably aren't considered HOF players (4th line players through "dynasties", etc. And should Cup wins be weighted differently for having played in a 6 team league vs 30?
Also, at some point you'll have to "eyeball" subjective points such as how many points to give for winning an award vs being in the final 3 candidates. Is a one-time winner better than a guy who was a final ballot candidate in 3 seasons? 4?
Dunno. Seems like an uphill battle.