Next up. My personal favorite template.

"That's my job. I'm the head coach. Let me manage that. It's my responsibility to explain to my players that this strategy is giving us the best chance to win."

...someone asked about "trusting your defense," as in, going for it hurts your defense's morale because it shows you don't trust them. Del Rio said let me take this one, and said something that's always stuck with me. He said something along the lines of...

Lastly, a story. I was on a panel with Jack Del Rio about 12 years ago. The 4th down topic was hot at the time. Over the course of the panel he went from downright hostile to skeptical to accepting. By the time we got to questions from the audience, ...

Second, what if you do go with the 2pt try first and succeed? Given the logic of the morale argument, wouldn't the players become even more hopeful and determined knowing they just need the xp?

First, if a team tries the xp first and manages the game to the point where time expires following a failed 2pt try, then it doesn't matter what their players' morale or intensity is like. They lose the game; end of story.

This is a fair point. Football players aren't playing cards or robots. They're people with beliefs and emotions. It's important to remember that. But 3 counterpoints, including a fun story:

Some kind of card game analogy might help convince the holdouts. Far from a perfect analogy but: In blackjack, would you rather see the dealer’s face-down card before you hit/stand? It’s going to be the same card either way, but of course the information is extremely valuable

Lots of different ways to organize and view the leverage numbers. Here's a mockup of a single team's rooting interest in all the weekend's games.

Our team put together an interesting way to look at 2nd order CFB playoff leverage. Here are some examples. For example, if you're a Vandy or a Miami fan how concerned should you be with the outcome of the Rice vs S. Fla game? (Plus all the other possibilities...)

Long live the 4th Down Bot Cc @KevinQ

One last note: this is all in a vacuum and treats a single play in isolation. When you draw the system boundary around the series, variance becomes crucial, especially downside variance. I think that explains why running is more common than a mean EPA analysis prescribes.

For example if the defense is too on guard vs the run, then the optimum is to keep passing. It’s complicated by the fact that each down-distance situation will have its own equilibrium.

But if the defense is leaning too far one way or another, the optimum is to keep spamming the opposite until they adjust.

Excellent thread. I think to really know solution to the run pass ratio question we’d to know the 4 “corners” of the payoff matrix. EPAs when off runs and def 100% expects run, when off passes and def 100% expects pass, and so on. Then we can solve for the equilibrium. But

@ProFootballTalk See for yourself! All the recommendations are publicly available. https://t.co/psVUDFkNTA

Do ESPN Analytics ever suggest anything but GO on fourth down?

MODEL LIKED PUNT BY A LOT

Huge No. 1 pick leverage today in East Rutherford. Chance to earn top pick in draft, per FPI Browns w/ win: 6% Browns w/ loss: 27% Jets w/ win: 8% Jets w/ loss: 27%

Happy 250th birthday, United States Navy. I'll celebrate by sharing the greatest recruiting ad ever

Alternate