This came to us over the weekend from Cold, Hard Football Facts troll Mark Wald, who takes us to task over our recent reports that "establishing the run game" is a meaningless pigskin aphorism with little foundation in actual on-field results. We thought it was worth sharing here.
He writes:
I love your site. I love your Quality Stats. I love most of what you do. But guys, your continued critique of the "establish the run" theory is slanted.
You make your own presumption about what "Hoge and Company" are saying, and then proceed to produce statistics that disprove the presumption. 
This is where your error is, fellas:  you've made the arbitrary decision that rushing yards or yards per rushing attempt is what "Hoge and Company" are using as the "establish the run" measuring stick.'s interesting that CHFF, on your own, is the Law when it comes to what "establishing the run" means. 
Newsflash: it's the number of rushing attempts that is important. The yards gained in total or per attempt is not as important. 
Want proof? I'll make an assertion. Take the games this coming weekend, or take the games from the beginning of this regular season through the present.
The team with the most number of rushes will be the winner in the majority of games. Now, I'm not about to disagree with Yards Per Passing Attempt. It's as important as you say it is.
But you'll find that Number of Rushes Per Game is every bit as indicative of winning football games as YPPA.
Now, I understand you made the statement in one of your articles that "any team can do something often, it's doing something WELL that counts." 
If the only thing we were trying to do is define what makes a successful or dominant running game, I'd agree with that assertion. 
But isn't the idea behind Quality Stats that they are stats that are indicative of winning? Well, Number of Rushing Attempts Per Game is one of the strongest winning indicators there is.
I have a hunch you guys know this. A site as excellent as CHFF, written by guys as on top of things as you, would not be oblivious to such a thing.
But for some reason, and I don't know why, you're not saying it.  You've chosen to, for God knows why, to focus on "rushing yards or average per rush vs. YPPA". 
In other words, you appear to have an agenda.  And that's disappointing, because that's what the "pundits" do.  
CHFF responds
Well, Mark, your assertion that "we're on top of things" is way off base, unless by "things" you mean recliners and futons and, on those occasions when mommy lets us out of the basement, a Big Wheel.
There's a lot of merit to what you say: maybe rushing attempts is as important as you suggest. We just haven't looked at it. Not out of bias, but because it seems like it would be more of an "after-the-fact" indicator.
Teams with big leads often crank up their rushing attempts to sit on a lead. It's not rushing that gets them that lead, though.
But maybe we're wrong. There's no reason we can't look into your theory more.
The one statement you make that we would whole-heartedly refute is the accusation of bias. We can assure you with 100-percent certainty that bias is not the source of our statements that "establishing the run" is overrated (or the source of anything ever published on this site).  
Our arguments simply came about by first looking at the data:
  1. Earlier this year we looked at all the best passing games in modern history (based upon YPA) and found that most of these teams were very, very good.
  2. Then just recently we looked at all the best running games in history (based upon YPA) and found that few of these teams were very good.
So our initial assumption that establishing the pass is more important than establishing the run was based solely upon what the evidence (in terms of YPA) told us. We had no idea what the data would tell us until we calculated it. If we had an idea in mind then went out to find data to support it, then bias would be a fair accusation.
But that's simply not what happened. We calculated data, and drew conclusions based upon the data. We can assure you that this is ALWAYS how we work. If the data told us something different, the story would have read much differently.
Then last week we got data submitted by reader Mike Stickles (which you reference) that looked at the importance of running the ball several different ways (but not in terms of rushing attempts, as you suggest). His data supported our initial findings. But if it refuted our initial findings, we would have published that, too. We think you'll see many times that we publish findings that contradict or refute our initial reports.
As for calling out Hoge in particular, it was only because he's one of these guys who analyzes every single game by saying they have to establish the run. We really have no idea what his definition of "establish the run" might be. It's just that he uses the term so often, it's almost meaningless. (In fact, just last night on ESPN he was asked to list ways New England can be defeated. Naturally, "establish the run" was on his list.)
He probably has no idea what he means, either, whether total yards, YPA, total attempts, whatever. To him, it just seems that every single game revolves around a team's ability or inability to "establish the run." And, so far, we've found little evidence to support that "establishing the run" will help you win football games. (As we noted earlier this month, if you look at the 30 most dominant run games in the history of the league, few of those teams even made the playoffs and just one won a championship.)
But we don't profess to have all the answers, either. We do profess to keep looking for them. So far, we've found little evidence to support the importance of "establishing the run." But if we do, we'll publish it. In fact, if you have this data, send it along. We will publish it, whether it supports or refutes our theories, all of which are built upon a mountain of statistical evidence.
- The Chief Troll