Skip to contents
This Is So Stupid

Some Notes On The Performance Of The Crud Meter

Image by Chris Thompson

In the wake of Tom Brady’s latest Super Bowl victory, Defector has received a steady stream of criticism from readers who spent the season receiving updates from the Crud Meter, which used proprietary analytical methods to determine precisely how crud Brady was during each week of the NFL season. We here at Defector usually prefer to let our work speak for itself, but given the intensity of the backlash that the Crud Meter has received, I now find it appropriate, as the architect of the Crud Meter, to push back on some of that criticism and clear up some misconceptions.

The most common critique that has been leveled at the Crud Meter is some version of, “Why did you stupid fucking dweebs jinx everything by spending a whole season trying to make fun of Tom Brady?” Though understandable, that response misunderstands some fundamental things about how the Crud Meter works.

Yes, the Crud Meter was invented to make fun of Tom Brady and his stupid noodle arm, but it also was the only analytical tool that realistically forecasted Brady, a huge ding-dong, and his stupid noodle arm winning the Super Bowl. The Crud Meter was always designed to account for significant errors in the weekly reporting of crud levels, and in doing so gave Brady a much better shot at the title than many may have realized.

Perhaps a lot of readers simply put too much stock into an early-season run of high-crud games, but the underlying numbers always showed Brady moving towards increasingly non-crud performances as the season went on. No, the Crud Meter never indicated that it was extremely likely or even definitely most likely that Brady would win a Super Bowl, but as far back as Week 9 the Crud Meter was showing that even just a ~3 percent error in crud readings would leave Brady with a 60/40 chance at having a 1-in-7 shot at winning the Super Bowl. Increase the range of possible crud misreporting to approximately 6 percent, and suddenly you’re looking at a 50/50 coin flip to give Brady a 2-in-8 shot at becoming world champion. You would not call either of those scenarios super likely to happen, but both would, analytically speaking, qualify as probably sorta likely to happen.

Given Sunday night’s results, it’s clear that the skewed crud levels that the Crud Meter anticipated were in fact present throughout the season. In that case, the Crud Meter actually provided one of the most accurate forecasts of Tom Brady’s chances at winning the Super Bowl, and those who insist on criticizing it are simply failing to grasp a basic principle of data journalism: The Crud Meter cannot fail, it can only be failed.