QBR (Total Quarterback Rating) and ESPN Brand Promotion

129027960_crop_650x440

Two years ago, to great fanfare, ESPN rolled out a new stat – total quarterback rating, or QBR. Apparently based on years of research and comprising thousands of lines of code, it was (and is) an effort to supplant passer efficiency rating, the now forty year old measure of QBs that has been controversial and problematic from the beginning. QBR brought a few advantages: 1) it was scaled to 100, which is easy to digest, in contrast to passer rating, with its weird “perfect” score of 158.3. 2) it is supposed to account for all of the contributions a quarterback makes, including running (and avoiding sacks), all of which is missing from the older measure; 3) and it is designed to account for context of various sorts. For example, a 20-yard throw in a blowout is weighted less than one in the closing moments of a close game.

QBR also had the initial advantage of being the brainchild of ESPN. This meant that it was likely to be flogged relentlessly by the World Wide Leader in its endless football coverage. But in the two years since it was introduced, QBR has had a rocky reception. Some in the stathead/analytical community have noted its considerable advantages over passer rating. Others, like Kerry Byrne and the folks at Cold Hard Football Facts have been more critical and note that passer efficiency correlates remarkably well with who actually wins and loses football games. In any event, as of a year ago, ESPN seemed to be downplaying QBR, especially because its complicated (and proprietary) formula was incomprehensible to much of ESPN’s own on-air talent.

But with a new season upon us, it appears to be back with a vengeance. On ESPN’s television channels (don’t we have a more up-to-date word for “channel”?), when sports center or other commentators talk about QBs, they seem to be referring exclusively to QBR to evaluate quarterbacks.

And this drives me crazy. If you want to put it alongside other stats, include passer efficiency, that would be tolerable. But let’s face it – among ordinary sports fans, almost no one knows how to make sense of QBR. And I don’t mean that they can’t replicate the measure to evaluate its validity. I mean that ordinary fans, when they hear a statistic, have some frame of reference to put it in. Why? Because they’ve been following sports for years and they have some sense of what good performance is, because they can connect it to previously established standards. 40 homeruns, 1000 yards rushing, 30 points in a basketball game, a hat trick, even a pass efficiency rating of 100, and so on. We have some idea of what these things mean because we have a context for them. Yes, QBR was scaled to make 50 average and 100 great. But in a larger sense, that doesn’t mean anything because it’s only two years old. We have nothing to compare it to.

To be clear, that doesn’t mean it lacks merit. WAR (wins above replacement) is a tool for combining all the measurable contributions of a baseball player into a single measure. It’s based on decades of research into what produces winning (and losing) baseball. And it was at the forefront of the debate last year over the voting for AL MVP. Mike Trout, the Angels’ 20-year old rookie phenom, blew the rest of the AL out of the water in terms of WAR. But Miguel Cabrera, the Tigers’ slugger, became the first triple crown winner in 45 years in 2012 and won the MVP award going away. To be clear, I think that vote was a joke. Trout was far and away the better player last year, but many voters (despite covering baseball for a living) did not like WAR, because it would involve doing some legwork they didn’t want to do, and were more comfortable relying on traditional stats that they could make sense of (and didn’t really have to think about). This is fine for ordinary fans, by the way, but a sign of professional laziness among too many sportswriters.

But here’s the point. As much as I think WAR is a superior measure of player performance than the hodgepodge of individual stats we still rely upon to evaluate baseball players, I would not want ESPN to project on screen graphics that *only* presented WAR. It won’t make sense to a lot of people, and I would want more information than that, if only because I have a larger mental library of other stats in my head with which to compare whatever players are being talked about. And WAR, to be clear, has a longer history and is better substantiated than QBR. And ESPN would never resort exclusively to WAR in comparing and evaluating players.

In its own way, this is ESPN at its worst. It’s pushing QBR, not because the measure has proved itself over time, and subject to lots of independent evaluation (and since it won’t disclose fully how it’s arrived at, this will be very difficult to do), but because it’s ESPN’s stat. ESPN and its stats guru Dean Oliver say they want to create more intelligent conversation around statistics. That’s great. But pushing QBR to the exclusion of other data that fans have some familiarity with is not helping us achieve that. ESPN is just asking, more or less, that we take at its word that QBR is the only thing we need to know about a quarterback. This is not serving ESPN’s audience. It comes across as just one more shameless form of brand promotion in the guise of a serious effort to better understand the games we’re watching.

(Update – August 12) – If you go to the homepage for ESPN fantasy football, only one statistical  chart is listed – the 2012 leaders in QBR. How many fantasy leagues even use QBR? What possible argument could ESPN make to justify featuring that statistic above *all* others?

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s