## Wed Morning QB: Did I like going for it on 4th down?

Since there is no game, this week I’m going to go back and comment on a facet of the Houston game that has gone unremarked upon. Twice on the first drive Dallas went for it on fourth down (on 4 &1 and 4&2 at the Houston 38 and the Houston 29 respectively). That suggests either (i) Dallas doesn’t have confidence in Buehler, (ii) Dallas thought it would be a high scoring game and they needed TDs, or (iii) something else (I’m hoping it’s the something else).

The options Dallas faced in the first situation.

1) Go for the 1st-down

2) Kick the field goal

3) Punt

What’s the best strategy? Statistics to the rescue. To help us analyze these choices we need the probability of conversion and the expected point values for each situation. For those who aren’t familiar with expected point values, here is a definition.

The value of the current down, distance, and field position situation in terms of future expected net point advantage. In other words, it is the net point value a team can expect given a particular combination of down, distance, and field position. First and goal at the one represents an EP near 6, while 3rd and 20 at a team’s own one yard line represents an EP of about -2.

For the 4th down conversion probabilities I’m using the data from Advanced NFL Stats here and for the FG probabilities and net punt yards I’m using the data from Advanced NFL Stats here.

For a 4&1 from the 38 the probabilities and EPVs are as follows:

-     4th & 1 outside the 20 is converted about 75% of the time

-     A FG from the 38 yard line (i.e. a 55 yard FG) is converted about 43% of the time

-     A punt from the 38 yard would net about 25 yards

-     EPV for a 1 & 10 at the 37 (assumes the 4th down is converted) is worth 2.84 points. EPV for Houston facing a 1 & 10 at their own 38 (assumes the 4th down conversion failed) is worth 1.31 points.

-     Points for a made FG are 2.3 points (3 minus 0.7 for the value of the ensuing KO). EPV for Houston facing a 1 & 10 at the 45 (assumes the FG failed) is worth 1.77 points.

-     EPV for Houston facing a 1 & 10 at the 13 (assumes 25 net yards on punt) is -0.04.

With the probabilities and EPVs we can calculate the net payoff for the various choices.

1. Go for it

(0.75 x 2.78) + (1-0.75) x -1.31 = 1.76

2. Attempt a FG

(0.43 x 2.3) + (1-0.43) x -1.77 = -0.07

3. Punt

0.04

We can see that Dallas made the correct call on the initial 4th down. Going for it yields the highest expected point value of 1.76. Surprisingly, punting has a larger expected point value than going for the FG. The expected point value for punting is positive since it likely leaves Houston with poor field position. On the other hand, attempting a FG actually has a negative payoff because of the possibility of giving Houston the ball at their own 45.

And what about the second 4th down attempt from the 29?

On a 4&2 from the 29 the probabilities and EPVs are as follows

-     4th & 2 outside the 20 is converted about 60% of the time

-     A FG from the 29 yard line (i.e. a 46 yard FG) is converted about 70% of the time

-     A punt from the 29 yard line would net about 17 yards of field position

-     EPV for a 1 & 10 at the 28 (assumes the 4th down is converted) is worth 3.47 points. EPV for Houston facing a 1 & 10 at their own 29 (assumes the 4th down conversion failed) is worth 0.85 points.

-     Points for a made FG are 2.3 points (3 minus 0.7 for the value of the ensuing KO). EPV for Houston facing a 1 & 10 at their own 36 (assumes the FG failed) is worth 1.2 points.

-     EPV for Houston facing a 1 & 10 at the 12 (assumes 17 net yards on punt) is -0.11.

With the probabilities and EPVs we can calculate the net payoff for the various choices.

1. Go for it

(0.6 x 3.47) + (1-0.6) x -0.85 = 2.08

2. Attempt a FG

(0.7 x 2.3) + (1-0.7) x -1.20 = 0.89

3. Punt

0.11

We see that Dallas again made the correct decision. Even though the probability of conversion has gone down due to the longer distance, going for it still offers the highest net payoff.

Now some people have concluded that the decision to go for it on 4th down twice on the first drive suggests that Dallas either doesn’t have confidence in Buehler or that Dallas was expecting a high scoring game. And that is possible. Both of those factors would make going for it the more attractive than the result of the generic probabilities. However, I’d like to think that Dallas was looking at the probabilities. With the surprise onside kick in the Chicago game and these fourth down decisions it at least suggests that someone at Valley Ranch is paying attention to the numbers. I could be wrong, maybe we won’t see another surprise onside kick or an increased propensity to go for it on fourth down. But I’m hoping it’s a taste of what’s to come.

One likely objection is that I’ve used generic probabilities that don’t account for the specific strengths of the Dallas Cowboys and their opponent Houston. I acknowledge that. I’m not suggesting that teams should go strictly by the numbers. I do think that teams (and fans) should be aware of the baseline probabilities and then make adjustments (assessments) from those baselines.

Nullius en verba.

Another user-created commentary provided by a BTB reader.

## Trending Discussions

forgot?

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

I already have a Vox Media account!

### Verify Vox Media account

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

We'll email you a reset link.

Try another email?

### Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

### Join Blogging The Boys

You must be a member of Blogging The Boys to participate.

We have our own Community Guidelines at Blogging The Boys. You should read them.

### Join Blogging The Boys

You must be a member of Blogging The Boys to participate.

We have our own Community Guidelines at Blogging The Boys. You should read them.