Michael Lombardi's Rush Attempts plus Completions Magic Number Is No Longer Magical

"One of the things I do every Monday morning when I come in here is I always look at . . . this is a Bill Parcells thing, you always look at rushes and completions, so you take the rushing totals, and you take the completed pass totals, and you add those two numbers together, it’s a stat that they utilize on Stat Pass, everyone has kind of integrated it into the analogy, and it helps you to determine execution of your football team. So if you have for example 30 completed passes and 25 runs, that’s 55, and the number to win is anything over 51."

First, before even looking this up, there is a lot of correlation versus causation going on here. Rushing attempts totals are highly correlated with winning. Since they are part of this equation, you’re going to get some correlation to winning here, based on that mere fact. I know it’s a nice easy number with simple addition that you can look to, but is it doing anything different?

Let’s start with this magic number thing. Last year, there were 18 teams that had exactly 51 combined rush attempts and completions. They went 11-7. Go up one more, and the teams went 11-11. At 53, they went 11-2. One below, at 50? 18-8.

Collectively, the teams with between 51 and 55 combined rushes and pass completions went 75-40 (65.2%). The teams just below the magic number, at 46 to 50, went a combined 77-48 (61.6%) in 2011. It doesn’t appear there is a magic number at 51, even setting aside correlation versus causation. It’s possible that this apocryphal number developed decades ago. Last year, teams that hit 62 or higher went 12-1. The only loss was the Browns in a game where they had 40 completions but less than 6.5 yards per attempt. Teams complete more passes today, and we also have more high tempo games, so that magic number doesn’t look like it’s so magic anymore.

And that brings up the next point. Combining these two to measure efficiency of the offense seems like looking at a beautiful painting with Joe Paterno’s coke bottle glasses, when you don’t need them. What happens when we look at the group of games from 2011 where teams had between 45 and 54 combined rushes and pass completions, but sort by relative percentage of the two numbers?

There is a reasonably strong correlation between percentage of pass completions, and losing. Perhaps we should instead say what we knew all along– rush attempts are highly correlated with victory.

The 40 most extreme pass completion teams in this range, none of whom had more than 23 rushes:

9-31 record, 7.25 yards per attempt, 65.2% completions, 3.9% td rate, 2.8% int rate, 5.1% sack rate, 4.18 yards per carry

The 40 most extreme rush attempt teams, none of whom had more than 15 completions:

35-5 record, 6.39 yards per attempt, 55.2% completions, 4.9% td rate, 1.9% int rate, 7.7% sack rate, 4.84 yards per carry

Now, is that a good measure of offensive efficiency? The QB’s in the extreme rushing teams had a sub-par yards per attempt, completion rate, and sack rate, but likely benefitted from good field position and playing ahead, which is why the td rate is higher and the int rate lower.

The quarterbacks on the heavy pass attempt teams passed it better, and the int rate and sack rates were still better than average. The difference? The pass completion teams gave up 30.9 points. The rush attempt teams, 14.1.

Thus, combining the two probably depends on how it is combined. 35 rushes and 20 completions? You are probably winning. Of course, if you just start out a game trying to run it 35 times, you probably don’t get there because you don’t pick up first downs and then fall behind, and abandon the run.

[photo via US Presswire]