I haven't run the numbers because I don't have access to the data that I need. But I'm guessing that if all possible changes to genetic makeup are considered, you've got like a one in one billion or worse chance of mutating into something that is viable. If you want to mutate into something more beneficial it's worse odds. It's happened too many times in too few generations for me to accept that it's merely random (yes, I know natural selection isn't random, but the mutations are according to the prevailing theory). Does that mean the theory is completely wrong? Probably not. In fact, it explains enough things well enough that I would say it's highly unlikely that the entire theory is completely wrong. I do think it needs to be adjusted. Does it occur? Probably so. Do we have an adequate explanation of the mechanism? Probably not.