It is fair to say that literally no one in Britain’s first, second, third, or fourth estates saw Leave’s victory coming in this year’s referendum, just as virtually no one on the American left (it would be redundant to add “and media”) can imagine Trump winning. But there the analogy stops.
In Britain, there was no empirical basis for Remain’s certainty; to the contrary, “[t]he polls consistently indicated that there was a very real chance that Britain would vote to leave. Polling averages even showed ‘Leave’ with a lead for most of the last month; over all, 17 of the 35 surveys conducted in June showed the Leave side with the edge, while just 15 showed Remain ahead.” Remain’s overconfidence was nothing more than arrogance and the blinkering effect that social media can have when one chooses to close-ranks and have only friends of the same political stripe. (A deadly mistake, I submit.)
By contrast, American polls have consistently shown scant possibility of a Trump victory. 538’s model has fluctuated between overwhelming certainty of Clinton’s victory and near-certainty of it; only once in the entire year, in the last week of July, have the chances of a Clinton victory been below 60%. And when Trump has had good polling weeks, prima facie? The surge has invariably turned out to be in states that he is already likely to carry, which benefits him nil. Trump could win the popular vote, but whether he loses California by one point or twenty, he still loses California.
To be sure, I agree that polls understate Trump’s true support; I said so on the podcast in August. If I were inclined to vote for him, I wouldn’t admit it to anyone, least of all a pollster! But they understate it by a percent or two—not by enough, and not where he needs it. They might understate it by two in Ohio, but they certainly don’t understate it by five in Florida or ten in Pennsylvania, and if you think Trump can win without carrying all three of those states, I’d like to see your map.
Like Charlie Cooke, I fluctuate between depression and fury that, in a year that could have seen the election of a decent conservative President, a threadbare plurality of the GOP party nominated Donald Trump, about whom I have said plenty already. I will presumably be voting for Evan McMullin, although I may yet revert to my original plan—”write in Carly Fiorina or Laura Roslin.” My point in this post, however, is not whether Clinton’s coronation is a good or a bad thing; it is simply empirical reality. My unhappiness at it does not change the numbers. And unlike Brexit, where the polls consistently showed that it was close and could happen, our polls consistently show Trump headed to the worst defeat since Mondale.