Here is link to the page that generated my ss:
https://www.pro-football-reference.com/ ... season.htm
It's annoying because when you hover over "Rate+" with your cursor, it mentions "See glossary for details", but when you go to the glossary, there isn't an explicit definition:
https://www.pro-football-reference.com/ ... ossary.htm
I think of it in terms of the problem that the OP I mentioned was trying to solve. Historically, someone like Graham or going back further Friedman, did way better than his peers as a passer. Brady or whoever you choose as a modern QB just isn't going to be that much better on any statistical measure. We see more "clustering" in modern times, so if you want to rescue the concept i.e. give the modern players at least a chance to do well, you have to do some type of normalization. Exactly what PFR does isn't clear to me, but they mention:
"we computed how many standard deviations away from the league average each player was"
which I think is on the right track for both using the "relative to peers" idea and normalizing so that a player in any era would have a chance to do well. You can search for the text above on the glossary page. Hopefully someone can explain it better than me.
I think it's really interesting - I may nominate him for HOVG.
P.S. This doesn't really solve the career issue that was the subject of the thread I reference since PFR comes up with a single season metric. I don't know how or if you can come up with some kind of career metric - all I think you can do here is see how many top 20 or top 50 or whatever single seasons a player had. Friedman has no chance here since his stats are unofficial, but Graham has 4 seasons in the top 20 and Baugh 3.
