Culture

Don’t Worry About the Democratic Presidential Polls


Exactly twelve years ago, on July 29, 2007, national opinion polls declared the front-runner for the Republican Presidential nomination to be one Rudolph Giuliani, the bombastic former New York City mayor. In second place, seven points back, was a retired Tennessee senator and actor, Fred Thompson. Languishing in third place, another five points behind, was the eventual G.O.P. nominee, John McCain. Over on the Democratic side, on the same date, Hillary Clinton led Barack Obama by nearly thirteen points. Everyone knows how that turned out.

Twenty Democratic candidates are set to debate in Detroit this week, as countless Democratic voters wonder, with knotted stomachs, whether anyone will emerge to defeat Donald Trump, in November, 2020. So what do the early polls tell us? I asked around and found an array of specialists firm in their beliefs that the polls are iffy. “These numbers are fun, but I wouldn’t put money on anything,” Lydia Saad, a senior Gallup research director, told me. “Historically, among Democrats, if you had to bet at this point, you’d do a better job betting against, than for, the front-runner.” Which can’t be good news for Joe Biden, who is ahead but who slipped after his shaky debate performance, last month.

Jim Messina, Obama’s campaign manager in 2012, didn’t mince words: “Right now, it’s just too bumpy. There are too many candidates. There’s too much back-and-forth. ‘Oh, the polling shows Joe Biden is the best candidate to win the election.’ And then, after the first debate, ‘Oh, Kamala Harris came up, and she can win.’ And all of it is just bullshit.” At this stage, he said, polls can offer indications of what might happen, but he wouldn’t take them to the bank. One problem is that so little is known about so many of the Democratic candidates. Another is that so few people are paying close attention. And then there is the fact that a Presidential campaign is a bruising, billion-dollar proving ground. No candidate sails to victory untested and unscathed.

Soon after Biden entered the race, in late April, polls showed him to be far ahead of a trio of senators from blue states: Bernie Sanders, of Vermont; Elizabeth Warren, of Massachusetts; and Harris, of California. But Biden’s initial lead over Sanders, of twenty to thirty points, has dwindled to an average lead of about fourteen points, with Warren and Sanders roughly tied for second and Harris close behind, according to an aggregation of national polls compiled by Real Clear Politics. The next-closest competitor is the South Bend, Indiana, mayor, Pete Buttigieg, who remains in single digits. No other candidate reaches even three per cent.

A significant portion of the support for Biden and Sanders can be attributed to name recognition. More than eighty-five per cent of Democrats and Democratic-leaning Independents had opinions about the two men, Gallup reported this month, and about seventy per cent saw each favorably. Only forty-nine per cent had an opinion about Julián Castro, fifty-three per cent about Buttigieg, fifty-seven per cent about Cory Booker, and sixty-five per cent about Harris, leaving lots of room for these lesser-known candidates to define themselves. Their challenge is to be noticed, positively. History shows that it can happen. “The Democratic Party tends to try to kill off its front-runners,” said Jeremy Rosner, a managing partner at GQR, a prominent Washington-based polling and strategy firm that works with progressive candidates. “We saw with the first debate how quickly things can change. One good performance, one scandal, one misstep. Right now, boy, my feeling about the Democratic field is: Who the hell knows? Almost all of them are really impressive, and in almost all of them I can see their liabilities and how they get defeated.”

Campaign ledgers are littered with candidates who once surged to the lead in a Presidential-nomination contest, only to fade. A short and incomplete list would include Joe Lieberman, George Wallace, Jesse Jackson, Chris Christie, Jerry Brown, and Jeb Bush. Yet, though polls may have limited predictive power early in the game, they are hardly worthless, John Sides, a professor of political science at George Washington University, told me. “There’s some signal in the noise. People polling in the double digits really do have different odds of winning than candidates polling at one or two per cent,” Sides said. “You can’t dismiss the polls out of hand as not telling you something informative. On the other hand, they’re not necessarily going to pick the winner at this point, nor can they tell us what changes might happen.” Four candidates are currently polling in double digits: Biden, Warren, Sanders, and Harris.

When Sides refers to changes, he means calamities that could befall a candidate, through a scandal or an origin story that doesn’t add up. Think of the Democratic front-runner Gary Hart and his alleged affair, during the 1988 campaign, aboard the good ship Monkey Business. Or Ted Kennedy, who challenged an incumbent President, Jimmy Carter, and led him in the polls, only to stumble when asked, by Roger Mudd, in a televised CBS interview, “Why do you want to be President?” In another interview, Tom Jarriel, of ABC, brought up the past: “Senator, you cheated in college. You panicked at Chappaquiddick. Do you have what it takes to be President of the United States?”

All of those moments, of course, preceded the rise of Trump. On July 1, 2015, two weeks after he rode down an escalator at Trump Tower and entered the race, he was polling at six per cent, according to the Real Clear Politics average. A month later, he was leading the field, with more than twenty-one per cent of the total. From then on, all the way to the G.O.P. nomination, Trump met outrage with umbrage and trailed for only an instant. He launched his campaign by calling Mexican immigrants “rapists.” He cast critics as unpatriotic losers and working journalists as enemies of the people. He mocked John McCain, a former Navy pilot, for being captured during the Vietnam War. He told supporters at a rally that he’d like to punch a protester in the face. He refused to release his tax returns. He boasted about grabbing women’s genitals. He denied the accusations of more than a dozen women who said that he had groped them. And he barely lost his stride.

During the 2016 primaries, Trump, in his inimitable way, solved the same sort of publicity challenge currently faced by this year’s Democratic candidates. He captivated audiences and kept the attention of the media by inventing belittling, yet memorable, nicknames for his fellow-candidates and turning, without shame, to what he liked to call “truthful hyperbole,” which fact checkers now more commonly identify as lying. (“It may not be good for America, but it’s damn good for CBS,” Les Moonves, who was then the network’s chairman, crowed. He reported, in February, 2016, that “the money’s rolling in and this is fun.”) As a matter of marketing and political science, the formula worked like magic beans for Trump’s chances. “The general rule is that voters don’t change their minds without information,” Sides explained, adding, about the current Democratic field, that “to get information in a twenty-three-candidate primary is largely a function of news coverage. To get news coverage is to do something newsworthy. It’s a very self-reinforcing cycle.”

For the strategists and operatives whose success depends on actual results, polls can be a help, a hindrance, and a distraction, sometimes all at once. Jen O’Malley Dillon, a former executive director of the Democratic National Committee and a veteran of five Presidential cycles, is managing Beto O’Rourke’s Presidential bid. O’Rourke drew huge crowds during his unsuccessful U.S. Senate race against Ted Cruz, last year, but he has struggled to reach even three per cent in recent national Presidential polls. O’Malley Dillon tries to be zen: “Polls are a snapshot in time, and, most of the time, they’re not completely accurate, so they’re inaccurate snapshots in time. It’s easy to get caught up in it, but all we can do is put our heads down, run our campaign, and build for our candidate.”

After Trump’s haphazard campaign stunned most pollsters and analysts, some Democrats raced to explore new methods of tracking fast-moving attitudes and inclinations. Messina, the Obama campaign manager, is a fan of a marketing tactic often used in the corporate world called “social listening,” which posits that meaningful signals may emerge in conversations on social media before polls pick them up. For a fee, campaigns can subscribe to tools that trawl the Internet for tweets, posts, or other media that mention a candidate or issue. Specialists then use human or artificial intelligence to figure out what the words and images mean.

But some campaign strategists and experts see limits to social listening’s effectiveness, not least in its lack of rigor. “In the same way you could walk into a pub anywhere in the United States and get information on something that might show up later in a poll, it can give you a hint,” Jiore Craig, a member of Jeremy Rosner’s team at GQR who runs the firm’s digital practice, said. “My first question is, ‘How are you doing it?’ There’s a right way to do it, there’s a wrong way to do it, and it’s not easy.” For good measure, Craig offered a familiar reminder to avoid drawing conclusions from Twitter, whose self-selecting users are “people who are more engaged. They’re reading the news. If something is blowing up on Twitter, that doesn’t necessarily mean it’s blowing up among regular voters. Not by a long shot.”

Even as Democratic candidates take the stage at Detroit’s Fox Theatre on Tuesday and Wednesday, for the second national primary debates, the focus of the ground game remains in Iowa, where Obama made his case, in 2007, through relentless retail politics and attention to local media. His team treated national polls mostly as an annoyance, instead betting that a victory in Iowa would change the game. And it did. Although Obama faltered a few days later in New Hampshire, his triumph in a largely rural, overwhelmingly white Midwestern state demonstrated to voters and donors elsewhere that he could win the nomination. Clinton improved her own field operation and the contest became a slugfest, but she never caught up, despite a national polling average that showed her, as late as January, 2008—on the eve of the Iowa caucuses—around twenty points ahead.

The Iowa caucuses will not be held until February, 2020, but the current candidates and their legions of foot soldiers are already in the field. In Des Moines, J. Ann Selzer is trying to divine who will show up and how they will caucus, as she has for more than thirty years. Selzer directs the Des Moines Register’s Iowa Poll, which has been remarkably on-target, perhaps most impressively in 2008, when it showed Obama winning comfortably by marshalling an unprecedented voter turnout, including tens of thousands of first-timers. “A lot of pollsters build a likely-voter model and they base it on what happened before. I poll forward rather than polling backward,” she said. Apologizing for a lengthy explanation that “will get pretty wonky pretty quickly,” she told me that her group gathers enough data to build metrics “that we’re weighting to known population parameters.” In 2016, FiveThirtyEight called her “the best pollster in politics.”

But even Selzer is finding it particularly difficult to get a thorough read this year. “There are too many candidates,” she said. “That’s a lot for likely caucus-goers to comprehend—the nuances of this person versus that person.” Given that few Iowans are firm in their choices at this point, she realized that asking people for their first choice, or even their top two choices, might not reflect the underlying strength of various candidates. So she changed things up. She began asking voters which candidates they were actively considering. “This gave us their footprint,” she said, and helped predict the rise of Warren and Harris after the June debates, in Miami.

When I asked Selzer whether we can know how things will turn out for the Democrats this cycle, she pointed to a Register analysis from earlier this month by Nick Coltrain and Tim Webber, who reported that the candidates are on track to appear at an astonishing three thousand events in Iowa before the February caucuses. “The candidates are going to spend millions of dollars and thousands of events trying to change how we see it right now,” she said. “Anybody who tells you they can look at the field now and know who’s going to get the nomination is kidding themselves.”



READ NEWS SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.