After a gusher of coverage about horse-race polls in which President Joe Biden is losing to Donald Trump, it's worth making note of a nagging asterisk: Trump keeps underperforming his polling in the Republican presidential primary. Here’s a quick breakdown of the first several high-profile GOP contests, comparing Trump’s polling margin over his nearest rival to his margin of victory.
Iowa caucus:
• Final 538 polling average: Trump +34 percentage points (over second place)
• Final election result: Trump +30 (over the second-place finisher)
New Hampshire primary:
• Final 538 polling average: Trump +18
• Final election result: Trump +11
South Carolina primary:
• Final 538 polling average: Trump +28
• Final election result: Trump +20
In his newsletter The Tilt, New York Times data analyst Nate Cohn downplayed the gap between the polling and the outcomes as not "especially large misses." But Cohn still offered several theories for those misses, which are worth a second look.
One of those theories is the possibility that undecided voters are breaking late and against Trump. This theory bears out in the South Carolina exit polls, where former Gov. Nikki Haley won 67% of the voters who decided "within the last month."
Some election analysts wondered if pollsters have now overcorrected for their previous sins of underestimating the so-called "shy Trump voters" (i.e., Trump voters who refused to tell pollsters about their support for him) of 2016 and 2020.
The other side of that coin is a theory Cohn and others have entertained: Pollsters are underestimating the anti-Trump vote. As Cohn explained of these early 2024 polls, "[A]nti-Trump voters simply weren’t as likely to take surveys as pro-Trump voters."
Cohn gave the anti-Trump enthusiasm theory an extra look:
There is one reason the anti-Trump turnout might have relevance for general election polling: It’s consistent with other data showing Mr. Biden with the edge among the most highly engaged voters. This could yield a slight turnout advantage, even in a general election. It may also mean that the current polls of all registered voters slightly underestimate Mr. Biden compared with the narrower group of actual voters.
This wouldn’t mean the polls today are vastly underestimating Mr. Biden, but it could make the difference in a close election.
One data point that has already become very clear: Trump is consistently losing the most dependable voters to Haley by roughly double digits. College-educated, suburban-type voters vote. Their tendency to get to the polls is solid, and in this election, they are already proving to be anti-Trump—even in the Republican primaries.
In South Carolina, Haley won 54% of college graduates (to Trump's 45%); and in New Hampshire, Haley won 56% of them (to Trump's 42%).
It's impossible to say why the current polling is overestimating Trump's advantage and whether that will bleed over into the general election. Polling itself is just one data point in the constellation of points we use to guide us as we approach Election Day.
Another data point, for instance, is fundraising, where Trump is getting crushed by Biden, and even got outraised in January by Haley.
But at the moment, it's worth noting that this cycle's polling has revealed a consistent flaw. And it can't be bad for Biden that Trump keeps underperforming Republican primary polling.
Democratic voters know Joe Biden is old and MAGA voters like to pretend that Trump isn't just as long in the tooth. Both men were old the last time we did this and the only thing that’s changed is Biden is now a successful incumbent, while Trump is busy juggling trials and indictments.
Campaign Action
