There’s a lot of stupidity running around in politics these days. But one of the most idiotic claims out there is that crowd size at rallies is more indicative of a candidate’s level of support than polls.
Fox News host Eric Bolling is the latest to take this approach to poll denialism: “Here’s why polls really shouldn’t matter or shouldn’t ever matter. You pick up the phone and you say, ‘Who are you going to vote for?’ That person on the other end of the phone says, ‘Well, I’m going to vote for Hillary Clinton.’ They’re not out there voting. The people who are getting out in the street and going to a rally, those are people who get up off the couch and go hear something, go say something.” (Scroll forward to about 5:47 in the video to see this.)
Well, yes, it’s true that the people on the phones aren’t out there voting, but neither are the people going to rallies. Why? Because it’s not Election Day yet. The earliest of early voting hasn’t even started.
People who go to rallies are more involved in politics and more motivated by a particular candidate’s message. They are probably more likely to vote than those staying at home on the couch, but many of those on the couch actually will get up to vote. Voting is generally a much less burdensome method of participating in politics than attending a campaign rally.
But more to Bolling’s point, rallies couldn’t possibly be more indicative of vote preference than polls. In a 2012 survey from Pew Research, 10 percent of Americans reported having attended a political rally or speech. Compare that to 58.6 percent of eligible Americans who voted in that year’s presidential election. Way more people will vote than attend rallies.
Pollsters talk to both enthusiastic rally-goers and couch potatoes because they try to achieve a representative sample ― meaning they try to interview a cross-section of Americans that roughly matches the demographics of the American population. Then they use vote history, ask questions of the respondents or use some combination of those techniques to determine who is most likely to vote. These are the “likely voters” that you hear about.
Bolling wasn’t done, though. When it became clear he was losing ground on saying rally size mattered more, he turned to a different attack on polls: “No one is saying the polls are wrong.” (Except that he did essentially say that.) “The polls are a snapshot in time. It’s like saying ... you take a picture of the weather and expect that to be the weather 82 days from now.”
He’s partially right again. Polls are a snapshot in time ― they measure opinion at the time they’re conducted. But no one who knows even the bare minimum about polls ever looks at one and says, “Oh wow, this is exactly what the numbers will be on Nov. 8.” Because we all know polls are snapshots in time.
“Since 1952 no candidate who is ahead in the polls 2 weeks after the conventions has lost the election.”
But what we are saying is that, at this point in the campaign, polls are getting more reliable as indicators of how the election will turn out on Nov. 8. As many have pointed out, since 1952 no candidate who is ahead in the polls two weeks after the conventions has lost the election. That doesn’t mean Trump can’t win ― but it means that barring a truly jarring event, the trajectory of the numbers point to a Clinton win.
While we’re at it, another common critique of polls is, “Everyone I know is voting for Trump.” Next time someone says that, ask how many people they know. A few hundred? Maybe a couple thousand? There were about 130 million people who voted in 2012 ― any one person is probably only talking to a handful of people who are very similar to them.
And if they say they know for sure how all (approximately) 130 million voters will vote? Well, I know a few pollsters who will want to talk to them. They could make lots of money with that skill, if they actually had it.
Yes, polls can be wrong sometimes. Trump himself pointed to a recent poll failure when he tweeted that “They will soon be calling me MR. BREXIT!” Newt Gingrich pointed to a 68-year-old poll failure in a tweet recalling the 1948 Harry Truman election. But that was too long ago to matter. And Brexit polls were off by an average of 6 points ― Trump is down by nearly 9 points nationally.
Trump, his campaign and surrogates would be wise to believe the polls when all of them say he’s down. Denial doesn’t win elections.