One thing that commentators of the Left and Right agreed upon after the September 10 presidential debate was that Kamala Harris had resoundingly beaten Donald Trump. Harris had been the aggressor, and Trump had gotten angry, routinely taking the bait and venturing into topics more beneficial to his opponent than to himself. But if Harris won the debate so handily, why haven’t voters clearly moved her way since?
Evidence that ordinary Americans’ assessments of the debate differed from those of the commentariat emerged within days of the proceedings. First, Reuters conducted a focus group of ten undecided voters. After the debate, one of those voters remained undecided, while the other nine split two-to-one in Trump’s favor, either pledging to vote for him or leaning toward doing so. This was a small sample, but it certainly didn’t support the view that Harris had won the debate decisively—or at all.
Then, about 36 hours after the candidates had faced off, Insider Advantage released a poll of likely voters in Michigan—the state that, based on results in the past two Trump elections, should be the easiest of the seven key swing states for Harris to win. The Insider Advantage poll, conducted entirely after the debate, found Trump up 1 percentage point in the state. The prior month, IA had found Harris up 2 points among Michigan’s likely voters. Since IA’s later poll was conducted so quickly after the debate—before much of the media spin campaign to assert that Harris had won—its results are particularly revealing.
In the days and weeks that followed, evidence continued to accrue that the debate hadn’t improved Harris’s fortunes or hurt Trump’s. By the afternoon of September 24, RealClearPolitics showed 20 post-debate polls taken in one of the seven key swing states (Pennsylvania, Georgia, North Carolina, Michigan, Arizona, Wisconsin, and Nevada) by firms that had polled the same state in August. Of the post-debate polls, eight had moved toward Harris, eight had moved toward Trump, and four hadn’t moved at all. But the shifts toward Trump had been more pronounced, as the 20 polls, on average, had moved 0.3 points in his direction.
What’s more, most of the polls moving in Trump’s direction are rated highly by FiveThirtyEight, which generally rates the polls moving in Harris’s direction as poor or middling. Emerson, FiveThirtyEight’s tenth-ranked poll out of 282, is the only outfit that polled all seven states both in August and again after the debate. It found movement toward Trump in four of those states (Pennsylvania, Georgia, Michigan, and Nevada), movement toward Harris in two (North Carolina and Arizona), and no movement in one (Wisconsin), with the seven states, on average, moving 0.4 points toward Trump. (Emerson currently has Trump ahead in four states, Harris ahead in two, and Nevada tied.)
In Pennsylvania and Georgia, FiveThirtyEight’s top-rated pollster, New York Times/Siena, found no movement between its polling in August and its polling after the September debate. But it found a four-point swing toward Trump in North Carolina (from down two points to up two points) and a ten-point swing toward Trump in Arizona (from down five to up five). Meantime, Trafalgar, the pollster that found the most net movement toward Harris, is ranked 279 out of 282 in FiveThirtyEight’sratings.
Since the debate, Harris has gained about one point on net versus Trump in RealClearPolitics’s national polling average. That small bump is hardly indicative of a conclusive debate win, and national polling doesn’t necessarily convey who’s winning this race, anyway—for good reason.
Of course, the debate is not the only event that transpired between August and today. Indeed, many of the previously cited swing-state polls conducted their August polling before the Democratic National Convention. Given those polls’ lack of decisive movement in her favor, that just means that Harris hasn’t gained much, if any, ground in those states despite the combined effects of her convention and the debate.
Given all of this, perhaps American voters viewed the debate far differently than did the pundits. In this issues-versus-intangibles election, Harris’s answers may not have made voters any more comfortable with her domestic policies. Many commentators agreed that her answer to the debate’s first question, on inflation and the economy, was her weakest of the night. It began, “So, I was raised as a middle-class kid” and included little on how she would combat inflation; members of the Reuters focus group complained about her “vague” answers on economic topics. Trump constantly brought up illegal immigration, another topic on which Harris had little to say.
Similarly, the candidates’ foreign policy exchange may have done little to persuade voters of Harris’s aptitude. The vice president emphasized, for example, that she “actually met with Zelensky a few days before Russia invaded.” Trump replied, “She was the emissary. They sent her in to negotiate with Zelensky and Putin. And she did, and the war started three days later.” Whether or not Harris actually met with Putin, or just Zelensky, this exchange was one of Trump’s most effective moments of the night.
On the intangibles, Harris’s prosecutorial style—more focused on needling Trump than on conversing with voters—may have played poorly with many centrists. Her tone and body language might have come off as condescending and could have made her seem less likable than many voters had previously remembered. By contrast, voters likely saw or heard little from Trump that, for better or worse, they hadn’t already seen or heard.
There’s a difference between a high school debate competition and a presidential debate, in which candidates vie for swing voters’ support. The commentariat seems to have forgotten this in scoring the Trump-Harris debate. Harris was indeed the aggressor, didn’t lapse into as much word salad as predicted, and clearly got under Trump’s skin. But that doesn’t mean that she won the debate in the eyes of swing voters looking for answers to today’s pressing policy issues.
Photo by Robert Nickelsberg/Getty Images