Archive for January 2013


Evaluating QBs: It’s All About Context

January 31st, 2013 — 12:13am

In part 1 of my Evaluating QBs series, we looked at what makes teams win and which of those things quarterbacks have control over. While wins can be useful to separate quarterbacks, that is only because they are correlated with the underlying factors that explain wins. Once we separate out and control for those factors, QB wins provide no further information.

Now that we have shown that QBs have some control over the plays they are directly involved in but no influence over other facets of the game–defense, special teams, and other offensive plays–we can now look at how many wins we’d expect each player to have based only on what they have control over.

We can get at this two ways: directly and indirectly. The direct way is to look at how often quarterbacks win based on their EPA (again, using Brian Burke’s Expected Points from Advanced NFL Stats). The indirect way is to look at how often quarterbacks win based on the EPA of everything else, what I’ll call “support”. That is, the sum of the EPA of the quarterback’s team defense, special teams, and non-QB offensive EPA. Continue reading »

2 comments » | descriptive, Evaluating QBs Series, Football, offense versus defense, player evaluation, talent distribution, team evaluation

Evaluating QBs: Why Not Wins?

January 27th, 2013 — 12:35am

Full disclosure: I’m a Peyton Manning fan. If you can’t get past that, stop reading now. Still there? Good, welcome.

Following the Broncos recent loss to the Ravens (and the subsequent Patriots loss), there has been a new wave of the old Manning vs. Brady argument. Clutch vs. choke. Winner vs. can’t-win-the-big-one. Add in another playoff loss for Matt Ryan and a couple big wins for Joe Flacco, and the debate is raging like never before.

If you’re reading this, you’ve probably at least touched on the subject this January. I have. The debate always seems to deteriorate into emotional arguments filled with snarky retorts and anecdotal “evidence”. Tuck Rule game is countered with the Helmet Catch. The Flacco Prayer is answered with the Tracy Porter pick six. And on and on. And on.¬†Every quarterback has been lucky, and every quarterback has been unlucky. Everyone can bring up some argument to support their claim. Without looking at the entire picture, we’ll never reach a valid conclusion. There has to be a better way.

A Clean Slate Continue reading »

Comment » | descriptive, Evaluating QBs Series, Football, offense versus defense, player evaluation, predictive, talent distribution, team evaluation

Back to top