What's Wrong With Shady McCoy?

With the unsurprising news that Brian Westbrook probably should miss a week or two, attention turns to his back-up, rookie LeSean McCoy.

And based on what we've seen so far, that's not a good thing.

We've been complaining so far about the Eagles' running game, but actually, it's not been the same for everyone:

The Eagles are actually tied for 11th in the league in yards per attempt, but as you can see, that's a number that's heavily influenced by the superlative contributions of DeSean Jackson.  If you take him out of the mix, that 4.35 drops to 3.73. 

But even that doesn't tell the whole story.  There's one running back who actually has a pretty decent per carry average ... the aforementioned Brian Westbrook.  And Leonard Weaver is doing fine as a short-yardage specialist.

Which means that for all the talk of a shuffling offensive line and maybe some poor gameplanning, one way of viewing these data is that the whole thing comes down to LeSean McCoy. 

And I do mean down.  His FO numbers look even worse:

Westbrook's numbers aren't great, certainly not up to his usual standard, but for a guy who was struggling to get up to game speed early in the year, they're not terrible.  If he qualified for the rankings, he'd be #14 right now in terms of DVOA. 

Shady, on the other hand, whew, that's some terrible stuff.  You can tell he's played more against some better defenses than Westbrook has (his DVOA adjustment is smaller than Westbrook's), but his rankings across the board are coming in the 30s.  That's bottom of the barrel stuff.

The biggest question, of course, is why.  The simplest explanation -- that he's just not very good -- doesn't seem borne out by the other evidence we have:  1) great college production, 2) good measureables, 3) doesn't look like he sucks when you're watching him.

Another possibility is that this is a statistical anomaly.  Possibilities there would include:  1) different defenses, 2) small sample sizes, 3) one or two big runs from Westbrook that are throwing off the mean.

To address issue one, we have the FO adjusted numbers above, as well as this chart:

Shady got the bulk of his carries in two games.  Against KC, he was reasonably effective.  Against Washington, he wasn't.  And you can take out the Washington game without really affecting the overall differences (it becomes 4.8 vs. 3.9).

There's nothing we can do about the small sample sizes.  But we can look to see if only a couple runs are skewing the averages.  Histogram (percentage of carries):

Graph form:

There seem to be a few issues here.  The first is that Shady's accumulating more negative runs.  We've seen a couple times this year when he's tried to reverse field against an entire defense -- which rarely ends well. 

On the other hand, if you're getting tackled in the backfield, that's generally a blocking issue.  But what's interesting is that Westbrook and McCoy have the same percentage of runs that go for less than one yard.  It's just that Shady gets more negative plays and Westbrook has more zero-yard stuffs.  My theory there -- assuming, again, these numbers are even meaningful at this level -- is that Westbrook might be just darting forward to take his medicine on those plays, while McCoy tries to make something happen -- and ends up going backwards.

McCoy also has a lot more one-yard runs (hole misidentification?), but the next big difference you see is really at the 4-6 yard level.  Those are runs that in almost any situation would be classified as at least a partial success.  Westbrook has been much, much better at accumulating those runs than Shady has.  In fact, if you group up those numbers a bit to get a smoother chart, you end up with this:

Pretty similar at the bottom and top, biggest difference in 1-3 vs. 4-6 (with the bulk of the 1-3 difference coming with one-yard runs).

So what does this all tell us?  It's by no means definitive, but the numbers certainly suggest that the issues here are related to the backs themselves, and not other factors.  That doesn't necessarily mean McCoy isn't a good back, but it might mean that he's not yet figured out how to run effectively in this offense.  And it also might mean that if Westbrook is out for a week or two, we need to start thinking about getting Eldra Buckley a few more carries.

Overall, I doubt Buckley can match up to McCoy in any way, but Correll Buckhalter demonstrated quite conclusively that a change of pace back can have success in this offense if he sticks to a handful of fast-hitting plays, doesn't think too much and just runs hard through the hole.  Might be worth seeing if Buckley can be the same guy, if McCoy doesn't get off to a fast start on Sunday.