I think you’re inventing a win condition that isn’t plausible and then assigning it to the other side of the argument. I’m not talking about recovering two onside kicks. I’m already declaring that a loss. Going for the 2 pointer early is still the generally correct strategy disregarding that.
Looking at the stats on the Cowboys/Rams game, the Cowboys scored their last TD with 2:17 on the clock. I don’t know what the timeout situation is, but I’m assuming for simplicity’s sake that if the Rams get the ball back the game is over.
So the Cowboys just scored a TD and are down by 9. They have to make a 2 point conversion either now, or after their next touchdown, in order to tie the game.
If they go for the 2 point conversion after the first TD and fail, the game is practically over, and they lose. Under a 1% chance of winning at that point.
If they go for the PAT, and then execute the two point conversion on the second TD and fail, the game is over, and they lose.
Either way, if you fail the 2 point conversion, regardless of when you take it, you’ve lost.
The former scenario has a slightly higher chance of success, simply because you have more information earlier about what you need (and conserve time for a potential third TD) but we’re talking about fractions of a percent chances of winning. This is the miniscule chance of the two onside kick recoveries that I’m willing to, declare an almost certain loss. Both scenarios almost always result in a loss. But if there is any advantage to be had, it’s by the earlier 2 point attempt. But this is what I dismissed at the top of my post for simplicity’s sake. I’m just saying that if there’s an advantage to be had here, it’s gained by going for 2 early.
So, then, why not take the one that gives you more options? If you go for the 2 point first, you can change your mind and kick a PAT if you get a penalty on the 2 point conversion attempt. Think of it this way: you get a false start on the 2 point attempt the first try. So you change to a XP and put off the 2 point for later, because having to go 7 yards instead of 2 is a bigger difference in win chance than choosing to go for the two early or late. But if you kick the PAT first, and go for the 2 second, and commit a false start, then you have to go for the 7 yard 2 point conversion, you have no choice. This alone I feel should be intuitive and compelling.
Additionally, you will go into the second TD with the ability to tie (XP) or win (2 point) the game, your choice. XP first only allows for a tie.
Or, let me try it another way way. I think this is an impractical argument, because the likelihood of winning in either case is so low, but I’m illustrating it to demonstrate the logic of it.
You have 2:17 on the clock. You just scored a TD. You go for a 2 point conversion. You fail. Now you know you need to get 2 more scores in order to win the game.
Or the second scenario. You went for the PAT. You drive down for another TD. There’s, say, 15 seconds left on the clock. You go for the 2 point conversion. You fail. Now you know you need another score - the same knowledge you had in the above example - except now there’s 15 seconds on the clock when you found out instead of 2:17, limiting your options.
Again, since both of these cases are a miniscule chance of winning, this is not the basis of my argument. I’m just trying to say that you don’t get an advantage by finding out later whether you succeed or fail in the 2 point conversion.
Here’s where you may be going wrong. It feels wrong and counterintuitive to you to basically lose the game at 2:17 instead of 0:15. “You gotta give your guys a chance until the end” is the intuitive response. But that’s not actually what’s really going on here. What’s going on is that you will need to attempt a 2 point conversion regardless of whether it comes earlier or later. If you fail either time, you lose the game. So failing at 2:17 is the same thing as failing at 0:15, even if it feels different.
Let’s say that going for 2 is a 50/50 chance, a coin flip, for simplicity. If you go for two, you flip the coin. It comes up heads. Great, you completed a necessary step towards winning the game. Or it comes up tails, you lose. But you have to flip that coin at some point, either this TD or the next one.
Waiting to go for 2 on the second attempt is basically flipping the coin and not looking at it until 0:15, when you score that second touchdown. That’s when you find out if you got heads or tails. It feels more intuitive to put off knowing whether you’ve lost or not until the end of the game, but it doesn’t actually make any difference. If it was tails, you lose whether you flipped it after the first TD and looked, or waited until the second TD to see what the result was.
The only actual differences work in favor of the earlier 2 point attempt. Being able to change strategies in the event of a penalty is a significant factor. Being able to go for 2 2-point conversions and flat out win the game if the defenses are worn down at that point and you don’t want to risk being on defense in overtime.
There’s no way in which going for the 2 point attempt second gives you an advantage. It feels more intuitive, because finding out whether you’ve lost or not comes at the end of the game instead of at 2:17, but it doesn’t help your chances to win.
This is sort of like the Monty Hall problem. It’s hard to explain the correct answer to someone who isn’t seeing it the way you are. Varlos is a poker player, and so am I - you have to be able to understand how to break up strategic decision into their game theory components. It gives you a different perspective on problem solving and how to separate the intuitive from the logical. This is the sort of decision that is simple for a poker player by counterintuitive for most people.