Finally finished my dorky QB rating metric. Jimmy surprised me a bit, finishing at 12th (I'm not fully amazed at this, but it's definitely the high end of the range where I tend to rank him).
https://www.49erswebzone.com/forum/nfl/196068-new-homegrown-rating-metric-jimmy-controversially-ranks-12th/
And Trey had a better rating, for the record (although with small sample sizes there will be large variations).
Each input in this rating is statistically tied directly to correlation with winning. The only divergence from this with rushing yards, which surprisingly has a very weak
negative correlation with winning (But it made no football sense to not include it, and because it is statistically insignificant, I flipped the sign and did a separate correlation analysis for guys who rush for a lot if yards per game, and gave them a bonus based on the
positive correlation with winning their rushing yards had.)
Other than that, I made sacks and dominant wins 25% on the QB.