I don't know a ton about the specific expected value from various drive starting points, but I would guess that there's a point where the expected points from each team's first drive tip in the favor of the team without the ball to start because of field position. And with the time it takes for 2 drives you'd be signing up to likely be ahead with only a couple minutes left at most. The prisoner's dilemma of one team picking spot and the other team picking ball would keep that advantage from being too extreme though, if you pick the 1 your opponent is gonna stick you there, but if you pick the 10? The 15? Probably closer to a coin flip(again, with the caveat that I know nothing about football probabilities) Yea, this sounds right. Probably would end up being highly opponent dependent, but if I'm facing Mahomes or Allen I'm probably chosing the 1 yard line and fully expecting them to take the ball. But we'd have zero data the first year. It would be great to have decisions with no conventional wisdom to back up the choice. I also think that it gives the other team at least some control over the outcome of the game, being able to choose the yard line. If they are too aggressive and get stuck with the ball and ultimately lose, its on the coach for making the wrong choice. Compared to a coin flip which is literally random luck and no human involvement.