Why You Should Take A Deep Breath Before Reading The Latest Trump-Clinton Polls
Hooray! It's that time of election season again, when (depending on whom you support) every single poll is cause for either panic or triumphantly punching the air.
Election Day, by the way, is Nov. 8. That's almost half a year more of hyperventilation over polls.
That sounds exhausting.
But to be fair, there is actual cause for excitement in the polling numbers these days. Polls have swung dramatically in Donald Trump's favor in a matter of weeks. Two new polls released over the weekend showed Hillary Clinton and Trump within the margin of error of each other. The reality TV star and former secretary of state are now within 0.2 percentage points of each other in the RealClearPolitics polling average.
Whether that makes you ecstatic or enraged, calm down. Here's what these first few general-election polls do and don't tell you.
Let's Count Grains Of Salt
The No. 1 rule to remember about polls: They're a snapshot of how people feel right now. They do not predict how Election Day will look. And there are dozens of ways that snapshot can be distorted at any given point in the campaign.
So if you're looking for grains of salt to take with the latest poll results, there are quite a few to pick from:
1) Clinton and Trump are in vastly different situations.
Trump has soundly defeated all his GOP opponents. Clinton, meanwhile, is fighting a two-front battle — one against Trump and one against a primary opponent who, while very unlikely to win the nomination, is still very popular.
That is a much worse situation for Clinton than Trump.
"Polls are probably measuring her at a down point in the race right now," said Cliff Zukin, a professor of public policy and polling expert at Rutgers University. "This is a weak moment to measure Clinton support because Bernie Sanders supporters are beating on her.
Meanwhile, Trump is alone at the head of his party. That gives him the same kind of polling advantage any other presumptive nominee would have at this point.
"If one party has their nominee wrapped up before the other, we usually see a bump for the nominee who's the presumptive nominee," says Patrick Murray, founding director of the Monmouth Polling Institute. "And that's exactly what we're seeing right now."
Consider that after John McCain locked up the nomination in March 2008, his numbers bounced above Obama's briefly. But as Obama got close to the Democratic nomination, he gained ground back away from McCain.
It's not that the current polls are wrong or mismeasuring how people feel. Rather, it's that the current state of the race may be pulling poll respondents in one direction or the other.
2) The big stuff hasn't happened yet.
Conventions and the first debate are both two big potential turning points in public opinion, Zukin says. If this is the start of a proverbial roller coaster ride, those are two predictable sharp turns.
And then there are the unforeseen issues: In mid-2012, Obama had to deal with the Benghazi crisis and Mitt Romney made his infamous 47 percent remark. In 2008, McCain chose a running mate who some say cost him dearly on Election Day (one analysis says Sarah Palin cost him 2 million votes).
Any number of crises, revelations and gaffes could wildly skew these poll numbers — and with two deeply unpopular candidates, there's plenty of reason to think either candidate could have his/her share of Swift Boat or 47 percent moments.
3) Late-May polling is historically bad.
Fun fact: Statistically, we are coming up on what should be the point at which polls are farthest-off from the eventual Election Day outcome. That's what Princeton University's Sam Wang found in a recent analysis.
"Truly, now is the single worst time to be paying attention to fresh polling data," the neuroscientist and polling expert wrote in a Sunday blog post.
With the election 169 days away, he added, we're almost at the time of the year where polls tend to deviate most from the eventual general election outcomes.
Past elections give all kinds of anecdotal evidence of this. Maybe the most popular example is Michael Dukakis' 10-point lead over George H.W. Bush at this point in 1988.
As for why polls should be the farthest-off from Election Day right now, Wang writes that he doesn't know. Quite possibly, it's because right now, parties are usually still going through their nomination contests. That would lend weight to the theory that Clinton's numbers could shift after she is decisively the Democratic nominee.
Once again, of course, polls provide good snapshots of how voters feel right now, but polls aren't meant to be predictive. And today's polls right now might be a lot farther from Election Day outcomes than those in coming months.
4) Unprecedented unpopularity
Clinton and Trump are the most unpopular candidates on record. And because they're such anomalies in that sense, Murray explained, it's hard to explain what polls are saying with much certainty.
"Because it's such an unusual and unprecedented situation it's hard to know what this will mean," he said. "Usually if there is a wide gap in favorabilities between the two candidates, it's no contest, but a small gap could go either way. But usually the small gap is with them both being positive, not negative."
"It's not a situation where we have any evidence in prior elections for how this will play out," he added.
But Still: Fewer Grains Of Salt Than Usual
Yes, there are the Dukakises who led earlier on, only to lose their footing. But this is not 1988, and Clinton and Trump are no Dukakis, for one glaring reason: Everyone knows who Clinton and Trump are.
In an ABC News/Washington Post poll released Sunday, 97 percent of Americans had an opinion on Clinton or Trump — that is, they were able to say they saw them favorably or unfavorably. That's unusual compared to many past election years. Usually, non-incumbent candidates post higher no-opinion numbers.
That ABC-WaPo poll in May 2012 found that 14 percent of Americans had no opinion on Mitt Romney. In 2004, 12 percent had no opinion of John Kerry. In 2000, more than 20 percent of Americans had no opinion of George W. Bush, and 20 percent said the same of Al Gore. For Dukakis in 1988, it was 25 percent.
And because people already know what they think of Trump and Clinton, it could be tougher to change those opinions.
"These are more likely to be an accurate assessment of a starting point than if the candidates were new," Zukin said.
What About Bernie?
To be clear, Sanders' path to the nomination is exceedingly narrow at this point. He would need to win 68 percent of pledged delegates (that is, 68 percent of the vote in the remaining states) at this point in order to win the nomination.
That looks unlikely, to put it mildly. In California, with its 475 pledged delegates, Clinton leads in polls right now by around 10 points.
But still, Sanders is performing better in general-election matchups against Trump than Clinton is — a fact he touts constantly as a reason he should still be in the race.
It's true that Sanders is viewed more favorably than Clinton right now. Multiple polls have indicated that her supporters like him better than his like her.
However, it's also true that Sanders simply hasn't been subjected to the kind of scrutiny and attacks that Clinton has. Clinton has been reserving her harshest jabs lately for Trump, not Sanders, and Trump has likewise been saving his attacks for Clinton. Moreover, Murray adds, Sanders benefits in those polls from Clinton and Trump's uber-high unfavorability numbers.
"You get this hypothetical situation in which you say, 'Give me anyone who's not Hillary or Trump,' and Sanders is the last man standing in that equation," he said.
Moreover, this happens in primaries. There is an emotional fight on the Democratic side right now, with Sanders supporters pitted against Clinton's. Once a nominee is selected, that can still change, or at least Clinton hopes — just as her supporters eventually came around to Obama in 2008.
Copyright 2021 NPR. To see more, visit https://www.npr.org.