Although Trump conceded earlier this month that “maybe crowds don’t make the difference,” he often brags about jam-packed rallies and some of his supporters follow his lead. “You look at a Trump rally and there’s 12, 15,000, 10,000 people and then you look at Hillary Clinton and you have, I don’t know, 1,500, 2,000,” Fox News’ Eric Bolling argued last week on “The Five,” undeterred by comments from his increasingly incredulous co-hosts.
Pointing to impressive rally crowds is a longtime tactic of candidates on both sides of the aisle when they’re hearing little good news from more scientific polls.
“There’s something going on in this country and the pollsters aren’t getting it,” Democratic presidential nominee Walter Mondale said in 1984, about a week before he would lose spectacularly. “Nobody who’s been with me for the last few days and has seen these crowds, seen their response, seen their enthusiasm, seen the intensity of their response and how they respond to these issues, no one who’s been where I’ve been, can help but believe that there’s something happening in this country.”
More recently, supporters of Sen. Bernie Sanders (I-Vt.) were prone to citing his ability to draw big numbers for his speeches.
To put this year’s crowd photos into perspective, we asked Americans in a recent HuffPost/YouGov poll whether they’d been to any rallies in 2016. We found out two things: Only a small percentage had, and even by that metric, Trump wasn’t clearly ahead.
Just 8 percent said they’d been to an event for Clinton, only 6 percent said they’d been to an event for Trump, and 3 percent said they’d attended an event in support of another presidential candidate. The same numbers applied to another pseudo-scientific gauge of support: 8 percent of Americans said they’re displaying bumper stickers or yard signs supporting Clinton, 6 percent said they have signs or stickers for Trump, and 3 percent said they’re expressing such visual support for another candidate.
But the vast majority of Americans haven’t done any of this. Eighty-one percent said they haven’t been to a political event this year, and 82 percent said their lawns and cars have remained politics-free.
Very few people, in other words, are shouting their electoral views from the rooftop.
And even those numbers probably constitute an overly generous measure of how many Americans are taking part in political activities. People like to think of themselves as politically aware, which can lead them to overstate how active they actually are. Research has also shown that the kind of people who sign up for online survey panels and participate in polls about politics tend to be more civically engaged than the average citizen ― although YouGov’s weighting process seems to make its results less susceptible to that effect.
A June 2012 Gallup survey found that just 12 percent of Americans had volunteered for a political campaign, donated to a campaign or attended a political rally. The venerable General Social Survey reported in 2014 that just 28 percent of Americans said they’d ever gone to a political meeting or rally. In contrast, nearly 60 percent of all Americans eligible to vote turned out in 2012.
The fact that only a tiny slice of the nation attends campaign events wouldn’t matter if that slice were a good representation of the entire voting public. Polls, after all, may survey 1,000 people or fewer. But while pollsters take care to make sure their results are representative of the electorate at large, the kind of people who show up at candidates’ rallies come from the unusually enthusiastic and committed end of the spectrum.
“People who go to rallies are more involved in politics and more motivated by a particular candidate’s message,” HuffPost’s Natalie Jackson wrote earlier this month. “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.”
That makes crowd size a less-than-useful tool for candidates trying to assess their chances of winning an election.
“Campaigns that rely too much on anecdotal evidence like crowd sizes when looking to measure progress are easily lulled into a false sense of security,” Kevin Madden, a senior adviser to 2012 GOP nominee Mitt Romney, told the Associated Press. “They assume, often wrongly, that the echo effect of being surrounded by big crowds of already converted voters is automatically translating into momentum.”
Most average Americans don’t think crowd size is a good metric, either. By a margin of 46 percent to 26 percent, they said how well a candidate is doing in the polls is a better measure of support than how many people are showing up to the candidate’s rallies. Overall, 55 percent of Americans said they believe Clinton is currently doing better in the presidential race, while just 25 percent think Trump is doing better and 19 percent aren’t sure.
Republicans are notably more enthused about the predictive power of big rallies: 40 percent consider them a better gauge of support than polls, while just 14 percent of Democrats and 27 percent of independents said the same.
That likely has less to do with an innate Republican mistrust of surveys (remember Mondale?) than it does with the fact that Trump is currently losing in virtually all national polls.
Although 30 percent of Republicans said they think Clinton is ahead in the race, 48 percent said they believe Trump is actually doing better.
The HuffPost/YouGov poll consisted of 1,000 completed interviews conducted Aug. 18-21 among U.S. adults, using a sample selected from YouGov’s opt-in online panel to match the demographics and other characteristics of the adult U.S. population.
The Huffington Post has teamed up with YouGov to conduct daily opinion polls. You can learn more about this project and take part in YouGov’s nationally representative opinion polling. Data from all HuffPost/YouGov polls can be found here. More details on the polls’ methodology are available here.
Most surveys report a margin of error that represents some, but not all, potential survey errors. YouGov’s reports include a model-based margin of error, which rests on a specific set of statistical assumptions about the selected sample, rather than the standard methodology for random probability sampling. If these assumptions are wrong, the model-based margin of error may also be inaccurate. Click here for a more detailed explanation of the model-based margin of error.