Skip to main content
A Project of The Annenberg Public Policy Center

Survey Says: Trump Misleads on Debate Performances


Railing against a Fox News guest who said Donald Trump won the presidency despite poor debate performances, Trump held up a series of opt-in online surveys to misleadingly claim that “polls” showed he “won every single debate.” Scientific polls showed otherwise.

In an extended riff that lasted more than 10 minutes during a rally in Colorado Springs, Colorado, on Feb. 20, Trump took issue with comments made by Real Clear Politics Associate Editor A.B. Stoddard, who cautioned viewers not to count out Democratic presidential candidate Mike Bloomberg even though the former New York City mayor looked “uncoachable” and committed “many blunders” in the Democratic debate on Feb. 19.

“I think that Donald Trump had disastrous debate performances. Many answers were so cringeworthy, you just couldn’t even believe he was still standing on the stage, and he’s president,” Stoddard said on Neil Cavuto’s Fox News program on Feb. 20.

Trump initially lashed out on Twitter saying that “I won every one of my debates,” and challenged Stoddard and Cavuto to “[c]heck the polls taken immediately after the debates.”

Trump expounded on that during the rally in Colorado a few hours later, again referencing the segment on Cavuto’s show but refusing to use Stoddard’s name.

“And I said, wait a minute, I won every debate, it’s true,” Trump said. “And we sent them polls. Poll after poll after poll. Not only won them, but I won them by a lot. … I’m just saying, every poll — you you know they do those polls right after 3, 4, 500,000 people. Time magazine.'”

Two minutes later, Dan Scavino, the Trump campaign’s social media manager, handed Trump a stack of news clippings. Trump then spent the next 10 minutes reading numbers from what he called “polls” taken by Time, CNBC, Drudge and others immediately after the debates that indicated the public overwhelmingly believed Trump won all of the primary and general election debates.

“I won every single debate,” Trump said. “Then three and a half, four years later, I have to listen to a person saying how we didn’t do well. Because these people [Trump then pointed toward the press] are among the most dishonest people anywhere in the world.”

We reached out to the Trump campaign for details about which polls specifically Trump was citing, but we did not get a response.

However, from the ones that Trump held up, we could see that he was referring to online opt-in surveys such as these from Time and CNBC, in which readers are asked to respond or vote online for who they thought won the debate. They are not scientific polls, and they carry disclaimers making that clear.

  • From a Time survey after the second Republican primary debate, published on Sept. 17, 2015: “DISCLAIMER: This web poll is informal, not scientific. It reflects opinions of site visitors who voluntarily participate. Results may not represent the opinions of the public as a whole.”
  • From a Time survey after the 12th Republican primary debate, published on March 11, 2016: “A disclaimer: Online reader polls like this one are not statistically representative of eligible primary voters. They are a measure, however imprecise, of which candidates have the most energized online supporters, or most social media savvy fan base.”
  • From a CNBC survey after an Oct. 28, 2015, Republican primary debate: “Disclaimer: This is an informal poll. Results are not scientific and may not necessarily reflect the opinions of the public as a whole.”

After the first general election debate between Trump and Hillary Clinton on Sept. 26, 2016, Trump tweeted the results of more than a half dozen of such informal polls, commenting, “Such a great honor. Final debate polls are in – and the MOVEMENT wins!” At the rally, Trump read the results from three of those polls — including two from the conservative outlets Breitbart and Drudge, both of which showed more than 75% of respondents thought Trump won the debate.

But those opt-in, online surveys stand in stark contrast to the result of scientific polls. Here’s a sampling of those polls on who won the first debate:

CNN/ORC International: Clinton 62%, Trump 27%.

Public Policy Polling: Clinton 51%, Trump 40%.

YouGov: Clinton 57%, Trump 30%.

Politico/Morning Consult: Clinton 49%, Trump 26%.

Echelon Insights: Clinton 48%, Trump 22%.

Reuters/Ipsos: Clinton 56%, Trump 26%.

NBC News/Survey Monkey: Clinton 52%, Trump 21%.

Gallup: Clinton 61%, Trump 27%.

Fox News: Clinton 61%, Trump 21%.

Washington Post-ABC: Clinton 53%, Trump 18%.

The statistical website FiveThirtyEight, which is owned by ABC News, wrote on Sept. 28, 2016: “Every scientific poll we’ve encountered so far suggests that voters thought Hillary Clinton beat Donald Trump in Monday night’s debate.”

Experts at the American Association for Public Opinion Research warn that opt-in internet surveys are notoriously unreliable, and they have cautioned that self-selection into a survey means that it is not randomized “making it impossible to know the probability or likelihood of any particular individual being included in a study.”

After several Fox News hosts cited online opt-in surveys to suggest that Trump won the debate, Fox News’ vice president of public-opinion research, Dana Blanton, sent a memo to producers and the politics team reminding them that such polls “do not meet our editorial standards.”

In the memo — obtained by Business Insider — Blanton wrote that “online ‘polls’ like the one on Drudge, Time, etc. where people can opt-in or self-select … are really just for fun.”

“News networks and other organizations go to great effort and rigor to conduct scientific polls — for good reason,” Blanton wrote. “They know quick vote items posted on the web are nonsense, not true measures of public opinion.”

“As most of the publications themselves clearly state, the sample obviously can’t be representative of the electorate because they only reflect the views of those Internet users who have chosen to participate,” Blanton wrote. “Another problem — we know some campaigns/groups of supporters encourage people to vote in online polls and flood the results.”

Indeed, The Daily Dot, which covers internet culture, reported after the first general election debate in 2016 that Trump supporters used sites like Reddit and 4Chan to bombard the online surveys and manipulate the results in Trump’s favor. Whether that happened, or to what degree, is impossible to know, but the fact that it could have happened points to the unreliability of such surveys.