I didn't think Donald Trump would run for president (even up to the day he announced). Once he announced, I didn't suspect his support would last very long. I was wrong, twice. And I've been wrong a number of times since.

In my defense -- and in the defense of those who, like me, were wrong -- skepticism was the sensible default position. After all, the moral of the "boy who cried wolf" fable is not "since he kept crying wolf, the villagers should have assumed that this time there was actually a wolf." What's more, the limited recent history of candidates who surge to a lead from somewhere besides the traditional political launchpad has been that they run out of fuel before leaving the atmosphere and plunge into the ocean, not that they enter orbit.

It's fun, for example, to point out how wrong columns like this one or this one or this one or this one or this one turned out to be. Given the information at hand at those moments, though, these were not insane positions. There was certainly evidence that Trump's support was more stable than past candidates that had seen such surges, but there was also evidence (and history) to suggest that it wasn't.

At some point, though, you have to come to terms with reality. I'm not sure when I came to terms with reality on this thing, but it was probably at the point last fall when I realized that Trump's campaign didn't look like a circa-2012 Rick Perry/Herman Cain bubble -- it looked like a circa-2012 Mitt Romney baseline against which others were rising and falling. (This was later than Head Fix-er Chris Cillizza came around.) And I'm still generally wary of Trump's ability to win a close race and think that the withering of his support right before Iowa and South Carolina are things that could be a good sign for his opponents.

But, for Pete's sake. The guy is in the driver's seat. He has "a stranglehold" on the nomination, in the unnecessarily violent wording of Cillizza. Again, there are reasons to be skeptical that he might win the nomination, as elucidated by Nate Silver over the weekend. There are not -- and for a long time, have not been -- reasons to dismiss Trump's candidacy out of hand.

Some of it, as here, is born of a desire to deny that Donald Trump will be the party standard-bearer.

(Kristol has, at least, been consistent.)

Some of it is just bad analysis, like when Joe Scarborough last week figured Trump had gone too far in criticizing George W. Bush's handling of the Iraq War. Or this one, suggesting that Trump would "peter out" at the polls after Iowa. You can see why someone might think these things; you can also see why they were incorrect.

Punditry is a tricky business that readers love. Huffington Post has a profile of a blogger from their site who was continually beating their staff writers on traffic, simply because his analysis consists mostly of "Bernie Sanders will win, full stop." People like to read things like this and to share them. They like the pros and the cons and they like to argue and they like to use these takes as jumping-off points on social media.

All you can do is try to offer the best insight you have at your disposal and cop to analysis you got wrong. And, I'll add, endure the slings and arrows of those who are amused at your having done so -- even when those attacks come from equally wrong pundits.

Donald Trump's deeply unusual campaign has meant a ton of incorrect analysis, and continues to do so. Not that he's the only one winning that fight. I looked at polling last summer and figured that Sanders was a long shot to win New Hampshire, despite what that candidate was claiming. That post was updated earlier this month with a single word.

"Oops."