[Ed-S: Bumped from diaries]
Who needs a football fix? Every year I like to size up the relevant QB competition to Michigan in an effort to convince myself that we’ll go undefeated. Please see the following links for the ghosts of seasons past as well as an overview of the general thought processes behind these projections (2010, 2013, 2014). I was unable to post diaries for 2011 and 2012. This exercise is pretty hit and miss but its fun so let’s get to it.
I don’t actually expect anyone to click through to all those links so here’s a summary of the foundational ideas I’ve developed over the years:
1. As far as I’m concerned, the feel good year end rating for a QB is about 140. It’s a tough standard but that’s what the subjective good looks like. Great starts setting in above 145.
2. I treat player skill as a ratcheting riding-a-bike type thing. You don’t just forget what you once “knew” When performance recedes, its because of other factors outside of skill. Therefore, performance must be parsed in order to not over assign skill to a given performance.
3. I strive for an accuracy of +/- 4 points in passer rating. I.e. 131 vs. 135 and 135 vs. 139 are acceptable but 131 vs. 139 is a miss. Below 130 is bad and if I put you in that category and you score at or below 130, I claim a hit.
4. Barring injury, if I pick the wrong starter the a priori assessment carries over to the replacement. More on this later.
5. I use Bill Connolly's RB Rating system to guide my commentary regarding specific players. I had developed a version but, the data streams I needed dried up and his method accounts for most of the factors I did so its good enough for me.
[Hit the Jump for a post-mortem of 2014 QBs]