What If the Polls Aren’t Wrong?

We’re in the final days of a crazy-making, year-and-a-half-long presidential campaign, so naturally some political writers are questioning our collective sanity.

Thursday and Friday saw several national media outlets consider the prospect that polling on the presidential race is wrong—that Republican nominee Donald Trump might have a much better shot at winning the White House than the press or polling firms anticipate.

“Could the polls be wrong?” asked CNN. “Do the polls have it all wrong?” questioned The Washington Examiner. “What if everyone’s wrong?” wondered Politico.

This was to be expected. The race has tightened of late, and there’s recent historical precedent for pollsters getting it wrong. There was Florida in 2012, where the polling average showed Mitt Romney ahead in the presidential race but Barack Obama ultimately won. There were screw-ups in the 2014 midterms, where Democrats underperformed polls. As we’ve heard ad nauseam, Brexit came as a shock. Polls are imperfect, as Tuesday’s results undoubtedly will show.

But this speculation is mostly manufactured drama. The fundamentals favor Hillary Clinton, and every major respected election forecaster has her winning, with probabilities ranging from 68 to 99 percent. As The New York Times notes, none of these forecasts relies on the same data, using everything from polls to statistical models to expert opinions to betting markets. For Trump to win, a whole lot more than the polls have to be wrong.

To be fair, the articles cited above all provide requisite context. Politico’s Shane Goldmacher describes Trump as “decidedly the underdog.” Kristen Soltis Anderson, the Examiner columnist and Republican pollster, says Clinton “appears overwhelmingly likely to become the next president.” CNN’s Richard Allen Greene balances speculation about skewing with skepticism. In fact, he and Anderson both rightly observe that Brexit polling was actually close, with Anderson citing this excellent New York Times analysis:

The 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.

Anderson is a welcome example of a clear-eyed Republican who recognizes the right’s skewed-poll obsession during the last presidential campaign for the lunacy it was. A millennial who studies her generation in the hopes of helping Republicans win their votes, she’s reflective of how establishment conservatives, chastened by their unfounded skepticism of four years ago, are largely accepting the polling in this year’s race.

“In the 2012 election,” Anderson writes, “the polls that had made Mitt Romney so confident that he was going to win were his own internal polls, based on models that failed to accurately estimate voter turnout. But the public polls, especially statewide polls, painted a fairly accurate picture of how the electoral college might go.”

Her gentle reminder, for anyone willing to listen, is that “polls are still working fairly well when you actually look at the numbers.”

Try telling this to the Trump campaign. Donald Trump has railed against polls throughout this campaign, repeatedly and falsely claiming they are “rigged” against him—while also touting unscientific online polls when they reflect favorably on him. His surrogates follow his lead, complaining on cable news that the polls are “skewed”—the most famous example being this exchange between Trump attorney Michael Cohen and CNN’s Brianna Keilar.

And on Friday, his campaign manager Kellyanne Conway, who is herself a pollster and has very occasionally taken a more fact-based approach to issues, was happy to tweet out Goldmacher’s piece—with a winking adaptation of his headline.

For Trump’s campaign, this attempt to discredit the polls is not merely intended to motivate his supporters—to convince them that he still has a chance to beat Clinton, as long as they turn out. It’s also a prelude to an even more delusional, and infinitely more destructive, attempt to discredit the election results themselves. After all, if Trump loses a “rigged” election on Tuesday, as all the polls predict, then the polls themselves also must be rigged. As always, there’s a logic to the man’s madness.