Posted by xfitwolf on 2/4/2014 1:37:00 PM (view original):
I read in some forum about a formula some use and just wanted some more opinions on it. IIRC, you need to see the player's ratings jump after his first season (this won't apply for first year players).
(Difference in rating from season 1 to season 2) * 2.5 + season 1 rating = good projected final rating guess
So if, for example, my guy shows a 56 contact rating in season 1 turn into a 65 contact rating in season 2:
(65-56) * 2.5 + 56 = 78.5 projected contact rating
Is this widely used and fairly accurate?
I know this is pretty hypothetical, but I'm going to do some thinking out loud: could this be used somehow to predict a player's regression also?
Example (Age 33-32)*2.5 + Age 32 rating = projection of lowest possible rating ? Obviously this wouldn't be helpful unless there was someway to figure out how much per year the rating would decrease, allowing people to map out when a player becomes no longer useful at the ML level, but I just thought I'd put it out there