HUFFPOLLSTER: Support For The GOP Frontrunner Might Just Be Overstated

EVIDENCE THAT TRUMP’S SUPPORT OVERSTATED, BUT NOT BY MUCH – Nate Cohn: “Ever since Donald Trump has risen to the top of the polls, Republican strategists have questioned whether those polls might be overestimating his support. There is evidence to support that theory…[in] polling data provided by Civis Analytics, a Democratic data firm founded by the 2012 Obama campaign’s analytics director, Dan Wagner. It showed Mr. Trump faring worse than in many recent public polls, but it was not enough to call his lead into question. An analysis of the data revealed that [Trump] fares best among voters who don’t regularly participate in primary elections. Nevertheless, he still leads the other G.O.P. candidates, even among the most frequent voters. The analysis also suggests that although Mr. Trump fares better among irregular voters, it’s not by an unusual amount. And some other candidates also do better with infrequent voters.”  [NYT]


Trump, Bush, Carson do best with infrequent voters – Most public polls draw samples of all adults then use respondents answers to questions about registration and intent to vote to narrow to likely voters. The Civis poll drew its sample from registered voter files which include actual voting history. Cohn: “That gives campaigns a powerful tool for examining the likely electorate. If Mr. Trump had a big advantage among unlikely voters, a poll using a listed sample — like the Civis data — would be the way to find out….Civis tried different methods to measure the effect of voter participation. All the methods showed the same basic story, with Mr. Trump faring better among irregular voters, but not by an unusual amount….Civis used a simple model to estimate the relationship between candidate preference and vote history — measured by the total number of elections respondents had voted in since 2000. It found that Mr. Trump, Mr. Bush and Mr. Carson all did much worse among voters who had participated in more elections.” [More on the survey and regression model at Civis Analytics]


Methodology comes with trade-offs – Via voter file sampling, Civis polls 757 voters who identified or leaned Republican, but telephone numbers numbers were not available for all registered voters (a percentage not disclosed for this survey) and in this case Civis called only landline phones.

Another voter file poll from Republican firm – The GOP microtargeting specialists at TargetPoint released another landline-only national sample of Republicans selected from voter files, in this case conducted with automated, recorded voice interviewing. TargetPoint found big increases since May in the percentage of Republicans willing to consider Trump, Carson and Fiorina. They computed a “derived” vote preference using the percentage considering each candidate and estimated Trump’s support at 18 percent, just two points higher than the Civis poll. [TargetPoint release, full results]

LUNTZ’S ‘LEGS WOBBLE’ AFTER FOCUS GROUP PRAISES TRUMP – Sam Frizell: “A flock of two dozen mad-as-hell supporters of Donald Trump agreed to assemble on Monday night in a political consultant’s office to explain their passion for the Republican frontrunner…Frank Luntz, a fast-talking Republican pollster who frequently appears on television and writes newspaper op-eds, urged them on…The group was not a representative sample of the Republican party, or early state voters, as all of them had been selected because they like (or love) Trump and live in Washington or its suburbs in Maryland and Virginia…At the end of the session, the vast majority said they liked Trump more than when they walked in.“You guys understand how significant this is?” Luntz asked the press breathlessly when he came back into the room behind the glass. “This is real. I’m having trouble processing it. Like, my legs are shaking.” [Time]

It’s just a focus group – Jonathan Bernstein: “Yes, it’s August, and there’s little political news to report on, but can we at least avoid hyping the results of voter focus groups?…We could presumably assemble a roomful of George Pataki or Lincoln Chafee supporters (yes, it would take some work), and they would probably like Pataki or Chafee even if we gave them good reasons to find another candidate….[F]ocus groups are also a methodological mess. The moderator brings his or her biases into the room. So do all of the members of the group.” [Bloomberg]

‘Pollsters dumbfounded by Trump’ – Ben Kamisar: “Polling experts agree on one thing when it comes to Donald Trump’s presidential run: They’ve never seen anything like it. The billionaire businessman’s dominance of the Republican presidential race is forcing experienced political hands to question whether everything they know about winning the White House is wrong. [The Hill]

ANOTHER REMINDER THAT EARLY POLLS DON’T PREDICT WHO’LL WIN – Tom McCarthy: “For nearly six weeks, in survey after survey, Trump has soared above the rest of the field by a double-digit margin. It’s a dramatic performance, one the candidate himself is clearly exhilarated by. Except when you overlay it with, for example, the arc of the early frontrunner in the 2008 Republican nominating race, Rudy Giuliani...Next to Giuliani’s lead, Trump’s lead looks like … a joke. Trump is having trouble cracking 25%, while for months at a stretch in 2007, Giuliani swanned around in the 30s. And yet Giuliani ended up winning not a single primary or caucus.” [Guardian]


UMASS-LOWELL FORSWEARS PRIMARY POLLS – Joshua J. Dyck: “You might want to stow that notion in the same category as the presidencies-that-weren’t of Howard Dean, Herman Cain, Rick Perry, and Mike Huckabee. Like the Donald, they all had leads in national polls before the actual voting started in their parties’ presidential nominating contests in recent past cycles….Polling primary races this early in the game is by and large harmless, but also by and large useless.  The reason for this is that voters are simply not paying attention at this stage, and so illusory surges, such as the one Trump currently enjoys, merely reflect the impact of news coverage and name recognition, and not well-informed preferences. That is why, at the Center for Public Opinion at UMass Lowell, we aren’t doing any presidential primary polling now and won’t be any time soon.” [Commonwealth]

HURRICANE KATRINA, THEN AND NOW: A decade ago, Pew found a wide racial divide on how Americans saw government response to the disaster – Pew Research: “Initial reactions to the government’s response to the crisis were starkly divided along racial lines. In a national poll conducted Sept. 6-7, 2005, a week after the storm made landfall, African Americans delivered a scathing assessment of the federal government’s relief efforts. Two-thirds (66%) said that “the government’s response to the situation would have been faster if most of the victims had been white.” Just 17% of whites agreed – most whites (77%) said the race of the victims would not have made any difference.” [Pew Research]

Today, a racial divide continues to exist in how New Orleans residents see the recovery – HuffPost: “Ten years ago Hurricane Katrina brought devastating damage and flooding to New Orleans and surrounding areas. The slow response by the federal government in the aftermath left victims and many Americans alike to question whether race played a role in the delayed aid, as victims were disproportionately black. A decade later, two new polls find that race continues to remain a powerful force in how residents have experienced recovery. While nearly half of New Orleans inhabitants say the city has mostly recovered from the storm, there is a stark difference between white and black residents’ views.” [HuffPost]

BIDEN ‘RUNS BETTER THAN’ CLINTON, BUT… – Quinnipiac University: “Vice President Joseph Biden runs slightly better than former Secretary of State Hillary Clinton against leading Republican contenders in the 2016 presidential sweepstakes, and has the best favorability rating among top Republican and Democratic candidates, according to a Quinnipiac University National poll released today.” [Quinnipiac]

-Brendan Nyhan: “None of the differences in Biden/Clinton support are significant! 2nd example of this.” [@BrendanNyhan]

Plenty of ‘good news’ for Clinton – Matthew Yglesias: “[The Quinnipiac] poll shows Hillary Clinton coasting to a crushing victory in a three-way race against Bernie Sanders and Joe Biden, winning 45 percent of the vote, compared with 22 for Sanders and 18 for Biden.” It also shows Clinton beating Jeb Bush, Marco Rubio and Donald Trump in head to head matchups and “winning in a landslide in scenarios where Trump runs as an independent.” [Vox]

This Week’s Polls

-Clinton continues to hold a commanding lead in Iowa. [NYT, Suffolk University]

-Democrats still see Clinton as their best shot of winning in November. [HuffPost]

-A second poll shows Bernie Sanders ahead of Clinton in New Hampshire. [HuffPost, PPP]

-Donald Trump continues to lead in New Hampshire with 35 percent; runner-up Kasich takes 11 percent. [The Hill, PPP]

-Catholics are more likely than not to agree with the Pope’s political positions. [PRRI]

-Americans continue to remain pessimistic that the Iran deal could bring peace with the country. [YouGov]

-Millennials really are suckers for “artisanal” goods, a new survey confirms. [Harris]

-A majority of Republicans say there are too many GOP candidates. [HuffPost] 

HUFFPOLLSTER VIA EMAIL! – You can receive this weekly update every Friday morning via email! Just click here. Enter your email address, and click “sign up.” That’s all there is to it (and you can unsubscribe anytime).

THIS WEEK’S ‘OUTLIERS’ – Links to the best of news at the intersection of polling, politics and political data:

-Sean Trende and David Byler present an interactive calculator that allows users to see how turnout and vote shares by race translate into 2016 presidential election outcomes. [RCP]

-Hillary Clinton received more endorsements from elected Democrats in August than Bernie Sanders and the draft Biden effort has received over the entire campaign. [Bloomberg]

-Clinton’s general election trial-heat numbers have been static since March. [Daily Kos]

-Matthew Dickinson explains why it doesn’t matter that Clinton scores low on questions asking if she’s “honest and trustworthy.” [US News]

-Nate Silver finds Trump continues to dominate media coverage. [538]

-Trump attacks the top Republican contenders, but they don’t attack him back. [538]

-Pew rounds up polls on what Americans say about immigration. [Pew Research]

-ABC News joins AAPOR’s Transparency Initiative. [AAPOR]

— This feed and its contents are the property of The Huffington Post, and use is subject to our terms. It may be used for personal consumption, but may not be distributed on a website.