Thanks for restating parts of the problem in a much clearer manner!

And yea, that article is why this problem is wreaking such havock on me, and I were thinking of it as I wrote the OP. I'm not sure why I didn't link it.

However, I still can't resolve the paradox. Although I'm finally starting to see how one might start on doing so: formalizing an entire decision theory that solves the entire class of problems, and them swapping half my mindware out in a single operation. Doesn't seem like a very^good solution thou so I'd rather keep looking for third options.

I don't think I understand the middle paragraph with all the examples. Probably because the way I actually think of it is not the way I used in the OP, but rather an equation where expectation must be equal to actual probability to call my belief consistent, and jumping straight there. Like so: P=E/2, E=P, thus E=0.

Hmm, I just got a vague intuition saying roughly "Hey, but wait a moment, probability is in the mind. The multiverse is timeless and in each Everett branch you either do recover or you don't! ", but I'm not sure how to proceed from there.