A couple of weeks ago, I talked about Pro Football Focus's data on eight-man fronts and how they seemed to show that the passing game did more for Doug Martin than the other way around. Of course, that was based on limited data from just one outlet but did tell a coherent story, at least.
Of course, in comes a new data set and we have some issues with our neat, coherent story. According to ESPN Stats & Information, Doug Martin faced a "loaded box" 40 times on first and second down, and the Buccaneers as a whole faced loaded boxes 58 times on those downs. That's 13.3% of the time, which ranked 13th in the NFL. A nice and respectable average number, which completely contradicts Pro Football Focus' 27th rank with eight-man fronts.
Another note there: Doug Martin did not fare well against eight-man boxes, producing just 3.2 yards per carry, which appears to be well below the league average at a glance. I'm not sure how significant that is, though. We're talking about just 40 carries, which can quickly be skewed by a few outliers. Such as, say, those runs for loss at the end of Martin's nearly record-setting day against the Oakland Raiders, when the Bucs simply started running him sideways and for significant losses to run out the clock.
The difference between ESPN and Pro Football Focus lies in how each group defines its statistics, no doubt. ESPN only looks at player within five yards of the line of scrimmage and at most two yards outside the outermost player on the line, so perhaps PFF looks farther back or has a wider definition of the box. Who knows? It's just clear that a direct comparison really can't be made.
Which makes the effort to discern anything from these numbers slightly frustrating. Should we look to Doug Martin's low ranking among PFF's eight-man fronts, or should we trust ESPN's numbers on loaded boxes? Which one is more accurate? Intuitively I would be more inclined to trust ESPN's numbers because they seem to fit casual observation best, which is that the Bucs had a very balanced attack with passing and running feeding off each other. But I would not be particularly surprised if PFF's story of deep coverage and eight-man fronts turned out to be the right one, either.
All of this comes down to a simple problem, really: neither outlet has shown that its data is actually meaningful, let alone how it could be meaningful. Which means that for now, we can just point at the pretty numbers and say "isn't that cool".