Thanks,
@drewkaree . Pretty sure the stress is more in the middle of the bolt and runs laterally. The steering cable pulls on the middle of the bolt now. So this is where I think the clevis is *more* material (as there is no thread). I agree the head is less, but there is no appreciable force up/down on it. And if I can get a clevis pin in from top to bottom, then the cotter pin (or the clip--which I think I'm more likely to try) on the bottom is just there to prevent the pin from 'hopping' out. The head of the clevis is only taking the force of the weight of the pin.
However, that said, certainly a fair point that it would be less secure than the nynut. Big question is whether we need as much as a nynut to prevent it from coming out. If I still have to have them upside down, I'll probably stick with the nynut way (not sure if I want to trust the cotter pin/clip as the sole point of failure--someone kicks it just right and the whole thing could fall off). But hoping without needing the extra length for the nut to go on I might be able to go back to putting it right-side up, which then means a double point of failure.
Does that make any sense?