Quarterbacks 2008

I've been working on my RB projections but today I came across a statistical slicing and dicing of the QBs' 2007 output on pro-football-reference.com, here. A lot of times this early in the season, I make a mental note that I found a neat article and then later I can't remember where. This was too potentially valuable to ignore.

Chase Stuart employed something similar to the defense adjusted value over average. (See Footballoutsiders.com) You'll have to read Stuart's blog to try get the details, but he starts by weighting yards, TDs and INTs to come up with an adjusted yard per attempt stat - which is basically a measure of a QB's effectiveness - how well they move the ball while scoring and not turning it over.

However, this is a stat of limited value. After all, the defenses faced by each QB vary in quality and this is where the DVOA style calculation comes in. He then uses the defensive rankings against the pass and weights the performance of each QB based on the strength of the defenses faced, creating another ranking - the rearview adjusted yards per attempt (RAT). I don't pretend to understand the meaning of the number Tom Brady is 2.9 and QBs like Eli Manning are negative. The numbers bear no resemblance to yards per attempt stat, but the point is it's a ranking that includes both a weighted YPA and a DVOA. That there is valuable stuff.

It's still not perfect, though, because it's still basically a stat on effectiveness. I mean Carson Palmer's RAT is .07 and David Garrard's is the second highest on the list - 2.57. Garrard is amazingly efficient, even accounting for the strength of the defenses he faced, but he's no where near the fantasy QB that Palmer is.

So here's what I did. Realizing that all things are not equal, I created a list of QBs using last year's attempts, yards and TDs. I want a QB that throws the ball a lot, gets his yards and scores, afterall. Using those stats to create my baseline rankings, I then used Stuart's RAT as justification for tweaking my rankings.

Do I have a final list? No, there's still far too much offseason. Not only that, I have to be sure I tweak based on free agent moves, past history and coaching changes, though it doesn't look like those things will have much affect at the top of the rankings.

As of now, what do I see?

-- Brady, Manning, Brees and Romo are still the top 4

-- I am surprised to see Matt Hasselbeck 5th on my rankings. Hasselbeck mocked me last year by putting up career numbers after I said he was basically an average QB. With an O line that is getting more suspect and the RB position in transition, I suspect they may continue to rely on Hasselbeck's arm.

-- Carson Palmer's RAT is surprisingly low. Though he's currently sixth by this ranking method, I have to admit that right now, I might be tempted to let Palmer get by me in the draft out of fear that he drops off this year.

-- Eli Manning's not looking so hot by this method. That said, I was still considering nabbing him as my QB2 in case he channels his post season performance into the regular season

-- Jay Cutler looks highly intriguing by this method, bolstering my growing enthusiasm for him this year.

-- Roethlisberger is still top 10 by this method, but not as high as I expected given his 32 TD performance last year.

-- Derek Anderson's RAT was negative. He faced some sugarplum defenses last year. Between that and the Brady Quinn specter, I'm putting Anderson into RB2 territory. I don't want to count on him to carry my team.