The myth of using the run game to get to third and manageable

Monday I looked at third and manageable and how often the Seattle Seahawks found themselves in reasonable third downs, as opposed to third and long. There has been significant backlash towards the run game since the Seahawks dropped the Wild Card playoff game to the Dallas Cowboys, giving Russell Wilson his first ever postseason losing streak, having lost in the Divisional Round of the 2016 NFC Playoffs.

In the comments to that article the idea of run-run was presented yet again as able to generate third and manageable, as one reader posted the following comment.

It’s simple

When you run to gain 4 yards on first down and 3 yards on 2nd down, you end up with 3rd and 3.

When you run to gain -2 yards on first down, holding on 2nd down, 8 yards on the second 2nd down, you end up with 3rd and 14.

I think the idea is that if your pass is incomplete you’re on 2nd and 10, where even a “bad” rush will give you 1-3 yards and you’re at 2nd and 8. A second incomplete pass vs. a second inefficient rush, and it’s 3rd and 10 vs. 3rd and 6.

The idea breaks down when you get a penalty or get a TFL.

Here’s the thing, though, running on both first and second down is more likely to lead to third and long than it is to lead to third and manageable.

Wait, what?

Well, I won’t bore you all with the math, but here’s a super simple table that breaks down the expected yards on a first down run (left side of the table) and the expected yards on a second down run (top of the table).

In short, if a team calls a run play on first and second down, it will result in one of four things:

• 3rd & 7+
• 3rd & 4-6
• 3rd & 1-3 or
• a First Down.

The red area of the table is third and long. The yellow area is a third and medium. The blue part is third and short. And the green is a first down and the offense gets to start over.

Obviously, those numbers in the table are really small, and the table formatter here is far from perfect, so here’s a summary of what the expected outcome will likely be based on these probabilities.

• 3rd & Long (7+): 24.88%
• 3rd & Intermediate (4-6): 22.64%
• 3rd & Short (1-3): 18.28%
• First Down Attained: 34.20%

Basically, the most common outcome of a first down run followed by a second down run is gaining a first down. However, the second most probably outcome is third and long. Third and long is actually more likely to happen than third and intermediate or third and short, amd that's before even taking penalties into consideration.

Now, this is simply a single piece of the puzzle, so throughout the offseason I'll be adding to the discussion through similar posts. Thus, by the time Training Camp opens in six months we should have the entire book of probability out there, which would allow someone to build a model based on the combination of all the various probability trees that are available.