I didn't think it would actually happen.

Sitting in an airport with middle-class Britons a week before the U.K. referendum, I heard far more support for leaving the European Union than for staying in. But heading into the June 23 voting, I couldn't quite believe it. I didn't think it would happen simply because things like this usually don't. The status quo is a powerful totem. People don't like jumping off into the unknown. As polls moved toward "remain" in the waning days of the campaign, I assumed that we were seeing the usual pattern: People flirt with the new, dangerous outsider, then come home and marry the familiar boy next door.

It turned out my anecdotal evidence from the airport did better than the polls. And way, way better than the betting markets, which as late as 6 p.m. in the Eastern U.S. gave remain an 88 percent chance of winning. Betting markets failed worse than polls, worse than a casual survey in an airport. They failed, because as the blogger Epicurean Dealmaker pointed out on Twitter, "Markets distill the biases, opinions, & convictions of elites," which makes them "Structurally less able to predict populist movements."