We’re practically raised to believe that debates impact election results in big ways.
Millions watch, candidates spend countless hours preparing, media can’t stop talking, and writing, about them.
They’ve just got to be game changers.
But they’re not.
After the first debate, I noted that Sen. Kamala Harris’s (D-Calif.) vote share in the polling averages jumped about 8 points, while former Vice President Joe Biden’s fell by around 6 points.
However, within weeks of that sugar high, Harris’s support returned to just above her previous level.
Since then, detecting debate impact has required a microscope.
The biggest movements after the second debate were a 2-point pickup for Sen. Elizabeth Warren (D-Mass.) and a 3-point decline for Harris. Others saw 0- or 1-point shifts in support, on average.
Results after the third debate were similarly resistant to debate bumps. We don’t yet have as many post-debate polls, but IPSOS and FiveThirtyEight collaborated on a large before and after survey that found the biggest vote movement was again for Warren, who again rose by just over 2 points.
No other candidate generated a boost of even that small magnitude.
Similarly, the number of voters considering each of the candidates barely budged. Warren had the biggest gain here too, but it was only 2.4 percentage points.
All-in-all, there’s less game-changing than we might expect.
Why?
First, most primary voters aren’t watching the debates. IPSOS’s data suggests that nearly 6 in 10 Democratic primary voters did not watch any of the debate, and only 15 percent stuck through all three hours.
ABC says 14 million people tuned in. That’s less than half the number that voted in the 2016 primary between former Secretary of State Hillary Clinton and Sen. Bernie Sanders (I-Vt.).
Not surprisingly, there are some real differences between those who watched and those who didn’t.
Former Rep. Beto O’Rourke’s (D-Texas) net favorables (percent favorable minus percent unfavorable) rose a striking 15 points among viewers, but didn’t move at all among those who failed to tune in.
Sens. Cory Booker (D-N.J.), and Amy Klobuchar (D-Minn.) and Mayor Pete Buttigieg of South Bend, Ind., would also be doing meaningfully better if every Democratic primary voter had watched the exchanges.
Former HUD Secretary Julián Castro, by contrast, seems fortunate so many missed seeing him. His net favorables declined by a whopping 23 points among watchers, but didn’t move at all with those who missed his attack on Biden.
Second, as we know from study after study, in field after field, changing people’s minds is hard. Most people, most of the time, look at evidence and find their preexisting perceptions reinforced, not their minds changed.
The fact that the exchanges were mostly positive may further lessen their impact. Decades of research tells us negative information is processed more quickly and more deeply than positive information.
The debate was hardly a love fest. But the relative absence of hard-hitting attacks may make it harder to change minds, though as Castro learned, attacks can also backfire badly.
Fourth, with so many candidates, each gets relatively little exposure. In the course of a seemingly endless three-hour debate, only Biden and Warren got more than 15 minutes of speaking time. The rest held the mic, and the camera, even less.
Voters in the early states will get lots more direct exposure to the candidates through other mechanisms.
Speaking of three hours, when exposed to so much, it’s hard for folks to remember who exactly said what.
Harris got kudos from pundits for her opening message to Donald Trump, but three hours later, how many people had the slightest recollection of it?
None of this is to say that debates don’t have a cumulative impact or that there can’t be a game-changing moment.
But we should stop approaching each one as if it’s the decisive event of the primary contest.
Mellman is president of The Mellman Group and has helped elect 30 U.S. senators, 12 governors and dozens of House members. Mellman served as pollster to Senate Democratic leaders for over 20 years and as president of the American Association of Political Consultants.