Apparently, NASA is testing an EM Drive, a reactionless drive which to work would have to falsify the law of conservation of momentum. As good Bayesians I know that we should have a strong prior belief that the law of conservation of momentum is correct so that even if EM Drive supporters get substantial evidence we should still think that they are almost certainly wrong, especially given how common errors and fraud are in science. But, my question is how confident should we be that the law of conservation of momentum is correct? Is it, say, closer to .9999 or 1-1/10^20?