I'm signed up for cryonics, but to be fair, I agree that it would be irrational if the odds of coming back were 1,000,000,000:1 or something like that. Most people value their lives very highly, but not infinitely.
Depends on the requirements of the 1,000,000,000:1 option. If it's at very little cost to you, and if the alternative is certain, permanent death, the rational choice is still to take the incredibly slim chance of coming back. Otherwise it does indeed depend on the importance you put on your own life.
Of course, in the case of cryonics, there is no measurable probability of coming back, so the point is largely moot.
Depends how you measure I guess. If you insist that you need to wait x years and then count how many people were brought back against how many weren't, sure, there's no measure...
But you can break the question down and look for evidence about each step, it's not going to be a precise measurement but you can at least start estimating your belief certainty (as well as introducing bounds -- I don't think anyone's justified in a 90% certainty of coming back at this point, but nor do I think there's justification that it's as low as one in a billion). I've always liked http://www.overcomingbias.com/2009/03/break-cryonics-down.ht... as one approach to turning the problem into multiple conditional steps.