Tim Tebow destroying ESPN’s quarterback metric

About a year ago I began itching to write a book called The Namath Conundrum, or maybe The Namath Paradox, or even the Namath Algorithm, or some other such inscrutable title that would help ensure that rather than selling hundreds of copies I’d sell dozens.

This would be a book about great football quarterbacks like Joe Namath whose prowess and pluck have never been properly measured statistically by the NFL. Go look at the all-time leaders in passing efficiency and what do you notice? That Aaron Rodgers is easily the best quarterback of all time. Now, that may some day turn out to be a defensible statement. He’s really, really great. He’s also won a Super Bowl. But when you go down the list, you quickly come upon some names of people who haven’t won a thing. True, this is just one measure: passing. It doesn’t take into account running, play-calling, leadership, any of the intangibles. But it’s absurd on its face. Is Philip Rivers really a better passer than Joe Montana?

The official rankings indicate that someone named “Matt Schaub” is the 11th best passer of all time. That Chad Pennington and Daunte Culpepper had better careers than, say, Roger Staubach. That Brian Griese was best Griese.

Let’s put a stake through the heart of this monstrosity: The ratings say Jeff George was better than Johnny Unitas.

Now, there’s an obvious explanation: The rules have changed. In the old days, when a receiver went out for a pass, the defensive back was allowed to tackle him, bite him, hit him with a lid of a garbage can. Some receivers went out for a pass and were never seen again. They’re still looking for Lance Alworth.

Somewhere along the line, and I’m way too busy to look it up but will simply declare that it was in the 1980s, the rules changed to open up the passing game. The NFL wanted more zip, more points, fewer 3-yard runs between the tackles. Defenders could no longer pound on the receivers or even look at them askance. Now, routinely, quarterbacks throw for 400 yards in a game, a stat that used to be considered off the charts. You can still “sack” the quarterback but only if you do it nicely. Pass rushers have to bring their lawyers with them into the backfield.

.Thus the quarterbacks of the current era have absurdly great passer ratings, while the old guys, like Terry Bradshaw, have numbers that would get them benched today. The guy who exemplifies this more than anyone is Joe Namath. His career rating is 65.5, which is sub-Grossman.

Now, it’s true, Namath only had a few great seasons and had more interceptions than touchdowns. You can build a case against him. His fame rests forever on the singular achievement of Super Bowl III, when, unforgettably, he guaranteed a victory by the Jets over the Colts, or perhaps it was the Packers. Namath then shocked the world by making it happen. Until then, no one took the AFL seriously (the AFL-NFL merge soon followed). The modern era of pro football really dates to that game. [Discuss.]

Obviously, football has needed a way to recalibrate passing excellence over the course of many decades and changing rules. We need a new algorithm. This was what got me interested in the book idea. I’d come up with a new formula and it would make Joe Namath look like a great quarterback again and not so much like he should have been nicknamed Bowery Joe.

Well, naturally, other people had already thought of this idea. Indeed, sports in general has entered the Moneyball era, with sabermetrics all the rage [here’s Michael Lewis recalculating basketball excellence]. Everyone wants a better measure of performance. So here comes ESPN this year with a whole new algorithm for measuring quarterbacks. ESPN calls it QBR. As I understand it, ESPN has spent several years and gobs of money analyzing every single play by every single quarterback, assigning value and virtue based on a complex set of criteria that can only be calculated with the computers being used by the Large Hadron Collider to discover the hypothetical Higgs particle.

There’s just one problem with ESPN’s new algorithm: It can’t figure out Tim Tebow.

According to ESPN, Tebow is terrible. He can’t play this game. He ranks 30 out of 34 quarterbacks this season. He hasn’t had as good a year as, for example, Rex Grossman, according to ESPN’s metrics. The system uses a scale of 0 to 100, with 50 being the hypothetical average performance for a quarterback, and Tebow is well below average at 36.3..

This is all scientific and precise. QBR measures passing, running, fumbling, sacks taken, audibles slurred, egregious sideline posing, questions dodged in post-game interviews, and all the other important factors that go into being a quarterback. And, as the WSJ recently noted, there are a lot of things Tebow can’t do.

The only thing he’s got going for him is that he’s winning, and winning in the clutch, and making defenders look as if they’re trying to impersonate a matador as he runs past them. He has turned the Broncos into America’s Team, and has performed so many miracles that the Achenbro has declared, “This is God’s plan to convert the non-believers.” But he’s a zero to the ESPN algorithm.

Sorry, ESPN, but Tebow is killing your new quarterback rating system. Maybe you need to add something in there about leadership, and making your millionaire teammates want to play harder rather than just collect a check. Right now, you got serious algorithm problems. So I guess I should write that book after all.

Joel Achenbach writes on science and politics for the Post's national desk and on the "Achenblog."

opinions

Success! Check your inbox for details. You might also like:

Please enter a valid email address

See all newsletters

Comments
Show Comments
Most Read

opinions

Success! Check your inbox for details.

See all newsletters