I really appreciate the intelligent work and content of your post. There are not very many posts that compare. You are one of my favorite posters. I do have a few questions though:
* The goodness of fit of your model is 80%. An 80% GOF would be on the low end in manufacturing. Please excuse my ignorance but this does not seem good to me.
* How do you account for flyers, are the high and low end week to week stats thrown out since they could skew the data significantly?
* Do Passing Efficiency numbers really correlate to QB efficiency numbers since they do not take into account the play called, opponent, weather, or pass receivers? Can these factors really be considered constant across all of the data?
* Wouldn't you say that poor passing efficiency numbers are less of an indictment of the qb and more of a measure of the entire passing offense?