Quote:
Originally Posted by SmootSmack
Don't worry about Randolph. He's Chase Daniel 2010-meaning he won't be drafted (I don't think)
So are you going to share how you got to these results?
|
The actual formula outputs, not yet at least. I am serious when I say I did not use any variables that were not accounted for in the original post. I used one regression for completion percentage vs. college starts, another for completion percentage vs. college wins, and a third for performance in QBs best year vs. performance in the worst year, expressed as a ratio (the value metric was completion percentage times the percentage of career attempts thrown in that year...McCoy for example scored 12.92, 16.47, 19.79, and 21.15 in his four years...appreciating value).
The next step is to try to develop a way to deviate the comparables from the lines of best fit. Like I said, the comparables listed here was just me eyeballing the outputs, no math involved. That makes it more arbitrary than I would like, but again, it's December, and accuracy in the general sense is a good start in my book.
I'll probably have to steal a similarity scores formula from somewhere, and that will alter the results just a bit.