Mm. Clinton's not the sort of person I want as President as anything except a candidate of last resort, but the stuff that the right uses to attack her is silly when it isn't an outright fabrication.
The thing that scares me about Trump is that I've had the same sort of reaction before, the whole, "Whoa, no, there's no way that they'll manage to get the nomination, never mind win the general election!" thing... for people who were subsequently elected.
A black swan event happens every once in a while. And sometimes we misjudge something to be a black swan event. But it's confirmation bias, we really remember the black swan events. Ironically because the black swan events are so memorable we even sometimes make the rules around the outliers and ignore the normal cases because they aren't as memorable.
Oh, certainly. I'm not denying that in the vast majority of cases such candidates
aren't elected. But then, the outliers are what we have to be concerned with. As Bauglir said, there's little point building models which predict routine outcomes when we are already deeply familiar with the hows and whys of those routine outcomes and the model is no more accurate than the simple assumption that all outcomes will be routine outcomes.
Trump's concerning because he's an outlier candidate who has survived and thrived far longer than such usually do, and because the mixture of conditions which result in the election of a fringe candidate are much messier and difficult to pin down than the same for a mainstream candidate--this is, after all, why it's important to study them carefully, because there's more to them than their flat popularity among the general population, their application (or lack thereof) of campaign strategies (ferex, the duck or punch approach to addressing issues) in the sealed environment of mainstream Republican vs. mainstream Democrat, the presence or absence of a notable scandal in their past, &c.