The conditions were why I reduced his chance to make from inside 48 from his career 90% rate to 75%.

What does his kicking % have to do with the 25% chance of scoring a TD

Excellent point. You will see this effect often when each coach is deciding to call TO near end of 1H. Obviously one side benefits more than the other, as it is a zero sum game (win prob always sums to 1.0).

If Kyle knew that Lions kicker was sub par from 46 yards, I believe he would hope they go for FG, miss it, get 8 extra yards in the process, and benefit from a big momentum boost for the home fans and team.

That seems to be the easiest path to get the game turned around IMO.

Of course stopping them on 4th and 2 also changes momentum, without the additional 8 yards and also with the risk that Lions succeed and run an additional 3-4 minutes off the clock, keep 9ers D on the field, kick a 90%+ FG or get a TD that would likely put them in < 10% win equity position.

If FG chance is < 50% it seems going for it is sound. If it is 75%+, I suspect kicking it might be better. I want to take a few minutes tomorrow and get the break even #s and get a more substantive answer.

The Campbell error is when people reduce complex decisions to simple rules of thumb. He looks at ev in the abstract with no situational contextualization and with no further thought. If a > b shove, otherwise no, doesn't matter what oponent is doing, has done, or might do. Never mind that he overrates the precision and robustness of his ev calculation.

Gaga math

The timeout is "worse" in a sense that it's obviously bad but couldn't have cost very much win equity

I mean it’s obviously hard to say, but I have to imagine what little win equity existed got obliterated when the only non-onside kick path to victory was taken away

It obviously doesn't. I was referencing my post 1341 in which I estimated his chance of making the FG at 75% even though he is 90% from that range career, allowing for the conditions. The 25% estimation of scoring a TD if going for it starts with 60% chance to get the first down, and then is estimated from routine NFL data as to TD rates 1st and 10 from just outside the red zone, allowing for the slight chance of TD on the 4th down itself. I used approximate marks not equal marks. I think it holds up pretty good.

I still say in this situation you kick a bloop middle ball, or a hard squibbler. With 10 men up and only one back, you now have a loose ball with 11 guys going head long for it versus one, with of course the up front line of 10 dropping back as well but surprised as to the situation. There are 50 yards in there with nobody in it, and the ball bouncing around irregularly. It's a abeutiful play and if I was Belichick's special teams coach he would give me a raise.

Sorry my fault I wasn't clear at all - the 25% chance of a TD I think is about right, but why do you think if they convert the fourth down it's either a TD 25% of the time or nothing 75% of the time? What % of the time do they take a field goal?

Yes I didn't cover that. I also grouped the 90% from inside 48 together but obviously at the outer limits of that range it is nowhere near 90% as you get longer within that range. And if they get some more yardage on the next set of downs the FG% is going up up up.

My main argument is that you don't introduce huge swing factors into the game when up big late ... as in eschewing a chance to go up 3 scores thereby risking giving opponent a possession to cut it to 1 score. 40% chance you are giving them that by going for it. Of course he wasn't really 90% to make that FG or even 75% (my adjustment factor) probably, maybe more like 67% given the top of the range of the stat and the conditions.

I wonder where Campbell's breaking point is not to go: 4th and 4?? I guess the season stats would tell.

I don't know what the live line was but was it around 75% when they went for it and what did it drop to?

I get wanting to decrease variance as a fav but there was a full 22-23 minutes left in the game and their defense sucks. Seems like his analysis of "our defense sucks" held up nicely. Their offense had done whatever they wanted to that point in the game. You say "up big late", there was a quarter and a half left.


If they kick the FG and he misses, everyone just says why wouldn't you let them cook like you did all season you weak ass bitch

All things equal of course reducing variance with the lead is the preferred choice.

But comparing kicking a FG as an equivalent to gaining the 1st down is so far off.

Besides he was more like 55% to make it from 46, and they were in the low 40s to convert. The variance difference is basically irrelevant.

4th and 3 is a different conversation.