The Washington PostDemocracy Dies in Darkness

Two-thirds of Clinton and Trump supporters had few close friends supporting the other candidate

Donald Trump speaks during a campaign rally on May 5 in Charleston, W.Va. (Brendan Smialowski/AFP/Getty Images)

Up until Nov. 8 at about 11 p.m. Eastern time, there was no shortage of conspiracy theories floating around on the Internet about how Hillary Clinton and the Democrats had initiated a thorough, Machiavellian effort to steal the election. Donald Trump himself was spreading rumors on Election Day, back when he thought he was likely to lose.

He didn't lose, of course, and the rumors have since flipped: Democrats are spreading theories about how Russians hacked the elections and so on. The winning side always won fair and square, according to the winners. The losers, however, recognize cheating when they see it.

Why? One theory — my theory — is that skepticism about the results stems from the fact that a lot of people don't really know anyone supporting the other candidate. If you live in a community where you and all of your friends backed Trump, if you go online to Facebook or Breitbart or Twitter and it's all Trump Trump Trump, you're going to quite naturally be skeptical when you see reports about Clinton doing well. And vice versa.

In our postelection poll, conducted with the Schar School of Policy and Government at George Mason University, we asked respondents how many of their close friends were supporting Clinton or Trump. Overall, the numbers were about equal: A bit over half of respondents had a lot or some close friends who backed either candidate.

But that's mostly because supporters of one candidate knew a number of other people who backed that same candidate — and not many who supported that candidate's opponent. Two-thirds of backers of Clinton had close friends who also mostly backed Clinton, and didn't have many close friends who backed Trump.

There's your bubble.

Not that people weren't interacting unhelpfully with one another, perish the thought. We also asked if people had gotten into any arguments over the course of the election. Somewhat surprisingly, only a third of respondents said that they had. Slightly more Clinton supporters said that they'd gotten into arguments than Trump backers, but it was more common among Clinton voters who had fewer friends that voted for Trump. Among Trump supporters, the numbers were flipped.

Given how contentious this election was, it's very safe to assume that a lot of those arguments weren't between Trump and Clinton supporters, but between Trump supporters and backers of Jeb Bush, or between Bernie Sanders fans and Clinton fans, or between fans of either candidate and a guy on Twitter with an egg avatar. There were plenty of arguments to be had if you were interested.

But again, my theory is that there wasn't enough conflict, that Americans largely found themselves in little isolated pockets of broad agreement. That makes a result like the one from last week all the more jarring, with skepticism first by half the country and then, in a flash of Florida ballots, by the other half.

Maybe more arguments would have made the election better, not worse.