Skip to main content
Was The Second Debate The Beginning Of The End For Donald Trump?

The “narrative” coming out of CNN’s Republican debate last week has been that Carly Fiorina notched another victory, in part by crushing Donald Trump. Here at FiveThirtyEight, we take a lot of shots at political media narratives, so it seems only fair to point out that, in this case, the narrative is right: Eight national polls of GOP voters have been conducted related to the Republican race for president since the debate, and they show a couple of clear winners and losers — Fiorina won; Trump lost.

Before we get to that, though, let’s be clear that we’re still talking about polling several months before any voting takes place; a lot can — and will — happen before the Republican nomination is wrapped up. And just as we said after the Fox News debate, it’s better to look at the aggregate of polls1 (that’s why I waited more than a week to write this). Finally, to control for house effects — a pollster’s tendency to find results that favor one candidate or another — I’m comparing post-debate surveys to the most recent pre-debate poll by the same pollster (i.e., Quinnipiac to Quinnipiac), as long as the “before” poll was conducted within a month of the Sept. 16 debate.2

Fiorina +6 +3 +5 +6 +12 0 +2 +7 +5.2
Rubio +3 +2 0 +2 +8 +1 0 +2 +2.3
Bush +2 +2 0 -1 0 -2 +4 +3 +1.2
Pataki 0 0 +1 0 +1 0 +1 +0.3
Christie -3 +2 0 0 +1 +2 +1 -2 +0.2
Cruz 0 -1 +1 +1 -1 +2 -1 0 0.0
Graham 0 0 0 -1 0 0 0 -0.2
Jindal 0 0 -1 -1 0 +1 0 -0.2
Kasich +1 0 -1 -1 0 +1 +1 -3 -0.2
Santorum 0 0 -1 0 -1 0 -1 -0.3
Paul 0 -2 0 0 +1 -1 -1 -1 -0.4
Carson +4 -5 -3 0 -5 +1 -1 +5 -0.4
Huckabee -4 +1 -2 -1 +1 -1 -1 -1 -1.0
Trump -5 -1 +2 -4 -8 -1 -6 -3 -3.1


Carly Fiorina: As I said at the top, she was the biggest beneficiary of the CNN debate. She is the only candidate to gain relative to her baseline in every post-debate poll.3 In an average of post-debate polls, a solid plurality of Republicans who watched the debate declared Fiorina the debate’s winner (Trump came in a distant second). Moreover, it’s possible that the swift press declaration of Fiorina’s debate triumph augmented her bounce. Remember, it’s not only the debate that influences public opinion, but the post-debate media spin as well.

Marco Rubio: He was the only candidate other than Fiorina who didn’t lose ground in any post-debate poll, and Rubio gained in all but one.4 He also placed third, behind Fiorina and Trump, for the candidate most voters said won the debate. Perhaps most importantly for Rubio, he is now nearly tied with Jeb Bush as the top “establishment” choice in the polls (candidates who have held elected office before).


Scott Walker: :(

Donald Trump: The polls clearly show Trump losing ground after his showdown with Fiorina. Even though a number of Republicans thought he won the debate, a larger percentage of Republicans said he did the worst job in it. Zogby was the only poll that had Trump gaining ground after the debate, and Zogby’s Internet polls earned an “F” in the FiveThirtyEight pollster ratings. One cautionary note, though: Polls after the first debate also found Trump falling, but that drop didn’t last.

Steady as she goes

Nobody else gained or lost more than 2 percentage points. Debates matter, except when they don’t. And in the case of this debate, most candidates’ polling was unaffected.

Check out all of our 2016 election coverage.


  1. Zogby did not list candidates who earned less than 2 percent of the vote. In those instances, we do not count the Zogby poll toward the candidate’s average.

  2. When a pollster hadn’t conducted a pre-debate poll in that time frame, I compared its post-debate survey to the Huffington Post Pollster aggregate done in the same mode (e.g., live interview to live interview) because a candidate such as Trump had seen a steady rise in his polls over the last month that wouldn’t be reflected in a poll taken more than a month ago. Jim Gilmore doesn’t appear in the table because he didn’t poll well enough to make either the main stage or undercard debates.

  3. This includes one gain of less than half a percentage point, which you can’t see in the table because of rounding.

  4. Just as in Fiorina’s case, this includes one gain of less than half a percentage point, which you can’t see in the table because of rounding.

Harry Enten is a senior political writer and analyst for FiveThirtyEight.