Do we have to spell this out?Shutterstock/bikeriderlondon

This article is from the archive of our partner National Journal

If Ben Carson is a poor debater, somebody forgot to tell Republican primary voters. In each of the first two Republican presidential debates, political media watchers panned the neurosurgeon's performance. His soft voice and soothing tone did little to command the stage, as he was silent for much of the debate while his more bellicose rivals commandeered the conversation. And when he did talk, his policy proposals—most notably a flat-tax plan based on the concept of tithing—failed to impress debate analysts.

Carson did not fail, however, to impress Republican voters, at least judging by a before-and-after assessment of polling. In late July, Fox News put Carson at 7 percent in national polling. In mid-August, after the opening debate, he jumped to 12 percent. (The Wall Street Journal/NBC’s before-and-after polling had him slipping by a point.) And in before-and-after polling of the second and third debates, Quinnipiac showed him posting jumps of 5 points and 6 points respectively.

By the third debate, the media had grown more cautious in assessing his performance (National Journal for round 3 basically gave Carson a  ¯\_(ツ)_/¯), but amid the pundit-poll mismatch, the takes came flying in hot: Washington’s elite, bubble-enclosed media—went the conventional wisdom about the conventional wisdom—was too far out of touch for journalists to understand what “winning” a Republican debate looked like, and so through their looking glass they’d gotten it exactly backward.

So does that mean that when the media names its “winners and losers” of Tuesday night’s Round 4, should it all just be dismissed as Beltway nonsense?

Not quite. A comparison of post-debate analysis and before-and-after debate polling reveals that the Beltway press’s assessment of the candidates does not always square with polled voters, but it’s not always wrong, either.

Take the case of Wisconsin Gov. Scott Walker, where the media analysis closely tracked the voters’ collective verdict. Walker was bland in the first debate, the media panned him, and his slide in Iowa continued. Ahead of the second debate, analysts deployed the oft-hyperbolic “make-or-break” label to his performance. Walker missed, and he broke—dropping out of the race within a week.

So if Walker left the pundits looking prophetic and Carson left them looking pathetic, where else did the press zig where the polls zagged—and where else did the two move in tandem?

Carly Fiorina: The former Hewlett Packard CEO dazzled in her debut. Low poll numbers relegated her to the “Kids' Table” debate, but there she stood out, getting the coveted “winner” label for her showing. And the polls seemed to agree: She jumped 4 points in before-and-after polling from Quinnipiac, 6 points in before-and-after from NBC, and 4 points among CNN’s respondents.

And at the second debate, she again stood tall, wowing observers with a forceful (if factually challenged) challenge to Democrats to watch horrifying scenes that she said were contained in the undercover videos detailing Planned Parenthood’s sale of fetal tissue. The polling there is more complicated. Post-debate, Fiorina again saw a bump in her numbers—and a surge to second place in New Hampshire. But those gains have since faded, and Fiorina now polls near where she did before—all without any notable gaffes or much notable at all in debate No. 3.

Ted Cruz: The Texas senator got strong reviews for his performance in the third debate, when he chided the CNBC moderators for attempting to turn the event into a “cage match”—calling it evidence of a liberal bias in the mainstream media. The media turned around and anointed him one of the debate’s top performers (alongside Marco Rubio). And it appears the voters agreed: Quinnipiac’s before-and-after analysis saw him jump from 7 percent in mid-September to 13 percent at October’s end.

Marco Rubio: The media’s verdict after the third debate was glowing. Rubio made Jeb Bush look foolish in their exchange over his voting record, he fended off a moderator’s critique of his tax plan, and he flashed the eloquence and charisma that has propelled him up the political chain from the start of his career. The polls agreed: Quinnipiac’s pre- and post-debate put him up 5 points to 14 percent in national polls, while Fox News saw similarly modest gains (from 9 to 11 percent).

Jeb Bush: After failing to impress in a “low energy” first debate, Bush took a hit in the polls. He dropped in polls by both Fox News and Quinnipiac after the second debate. And in the third debate, when he had pledged to come out swinging but was remembered mostly for a schooling at the hands of Rubio, he tumbled again. It didn’t take a genius to call out Bush as being on the losing end of either round, but the media got it right.

Donald Trump: So if the pundits were correct (at least in spots) in their diagnoses of Cruz, Rubio, and Bush, have they been getting too much guff for their blind spot when it comes to Carson? Let’s not be too hasty, because in the debates and beyond, there’s no candidate for whom political media predictions have missed as badly as they have with Trump. Six months ago, he was dismissed as a joke candidate. When he rocketed to the top of the field, pundits promised his popularity would suffer for saying Mexico was sending “rapists and drug dealers.” When that fall failed to actually happen, he was “surely” in trouble for mocking Sen. John McCain for getting captured during the Vietnam War. Then it was his face-off with Megyn Kelly in the first debate.

The list of media-issued campaign death certificates issued goes on, but after tuning out the noise, the voters’ verdict has been clear and consistent. Trump has been at—or at least very near—the top of the GOP field now for six months. Certainly, he may tumble eventually, but at this point, the journalists who correctly predict the timing of that one will look less like seers and more like stopped clocks who just happened to be right.

So what’s actually going on here?

In deference to humility, and given how deep we are into an article about how often the media gets it wrong, hopefully you’ll excuse us if we pass on drawing any absolute conclusions. Instead, here are a few possibilities for the media’s hits and misses in debate assessment.

One: Calling someone a winner in a debate does not necessarily mean predicting he or she will rise in the polls. After all, these candidates aren’t up on stage auditioning to become Mr. October—they’re looking to win their party’s primary. And at this stage, that may be less about getting a short-lived rise in the polls and more about impressing mega-donors (or just donors) who have the funds candidates need for the furious stretch run at the start of next year.

Two: If there’s a trend to be drawn, it’s that the Washington media does its most accurate work in assessing establishment candidates. The predictions were spot-on with Walker, appear to be nailing it thus far with Bush, and have decently predicted the paths of Rubio and Cruz. With nontraditional candidates like Carson and Trump, the press has been a mess. But is this all that surprising? Being a political reporter in Washington means spending a lot of time talking to members of the political establishment. That gives one a decent feel for assessing how mainstream Republicans would assess mainstream candidates. It’s useless, or even detrimental, to assessing how people on the party’s periphery of politics would assess candidates whose strength comes from their status as entirely outside the party.

Three: There’s a bit of a Heisenberg problem here: The media’s declaring a winner or a loser may (depending on how much clout you believe we carry) affect who actually ends up winning or losing. Part of the point of a primary, after all, is for a party to pick a candidate who will win the general election, and so it makes sense that voters would be more inclined to throw their support behind someone they see as a winner—or whom they’re constantly told is a winner. Why else, after all, would campaigns send their surrogates to the “spin room” after the debate to breathlessly declare their client as the big winner? “The narrative” of political journalism is partly imagined (created by a mix of what’s happening and by how the mass media choose to interpret what’s happening), but that doesn’t mean it’s imaginary.

Four (and we promise this is it): Finally, and we apologize here to any good folks who may have paid to run advertisements next to this statement, but maybe it’s all just chaos. This is, after all, an inexact assessment of media coverage compared against an incomplete sample of polling. And all of it is describing a race in which the first vote (still!) won’t be cast for months—leaving plenty of time for voters to change how they feel about the candidates. The complications layer on: A debate performance could win a candidate gains as the “second favorite” of tons of likely voters, but none of that would show up in polls of the national horse race; outside events (like a rival dropping out or an unflattering past revelation) can move results far more than a poor debate performance; the universe is a place of infinite complexity; and as Kentucky reminded the nation last week, maybe the pollsters are getting it all wrong anyway.


Emily Schultheis and Zach Montellaro contributed to this article

This article is from the archive of our partner National Journal.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.