Every four years, the pundits say “it all comes down to turnout.” The line is such a cliché — it’s never inaccurate but also means nothing — that it’s sometimes a joke on social media.
This year, though, I’m not laughing. This time, the election really might come down to turnout.
Usually, this is not my view. If anything, I have tended to think of turnout as a somewhat overrated factor in explaining election results. The makeup of the electorate is relatively predictable; the big question is usually whom those voters will support. In Pennsylvania in 2022, for instance, the same voters backed a Democrat for governor by 13 percentage points and another Democrat for Senate by five points, and backed Republicans in the U.S. House. This wide variation all came from the same turnout!
But this election seems different. As we’ve reported all cycle, Democrats excel among high-turnout voters, while Donald J. Trump is strong among relatively low-turnout voters. He’s made his biggest gains among low-turnout demographic groups like young men and nonwhite voters.
This pattern has held all the way to the final stretch.
In the last wave of New York Times/Siena College polls of key battlegrounds, Kamala Harris led collectively among voters who turned out in recent primaries or the 2022 midterms, while Mr. Trump had a 12-point lead among the 2020-but-not-2022 vote and a 19-point lead among those who didn’t vote in 2020 (but who were registered at the time; new registrants are evenly divided).
This is an extraordinary change from as recently as a decade ago, when Democrats were presumed to be the party that benefited from high turnout. During the Obama era, Democratic chances seemed to depend on mobilizing young, nonwhite and infrequent voters to the polls.
Now, all these familiar maxims have been turned upside down. As the prominent Democratic pollster Anna Greenberg put it: “The Harris coalition rests on the most reliable voters (older, college educated). Trump needs every single low-propensity less educated young person to come out and vote for him.”
Ms. Harris can probably take it to the bank that primary-or-midterm voters will turn out in a presidential election. Those who didn’t vote in 2022, let alone 2020, are a little more iffy. Tens of millions of them will undoubtedly vote, but just how many — and exactly which ones — can easily decide the election.
In the extreme, the Times/Siena data suggests Mr. Trump could win the presidency, perhaps even fairly handily, if he could turn out all registered voters. To take a striking example: Ms. Harris or President Biden has never led a Times/Siena poll of registered voters in Michigan so far this cycle.
If, on the other hand, Ms. Harris could replay the midterm electorate, when more casual voters stayed home, she could easily win over 300 electoral votes and carry the swing states by a comfortable margin.
The broader demographic story of the election may also hinge on turnout. As we first reported a year ago, Mr. Trump is faring surprisingly well among young, Black and Hispanic voters, but almost all of that strength is contained among those who sat out the midterms.
This is not simply about education: Even the college graduates who sat out the midterms were far likelier to say they backed Mr. Trump.
Of course, just because Mr. Trump leads among irregular voters does not necessarily mean he will win the irregular voters who decide to show up. In the midterms, Democrats managed to draw a disproportionately Democratic group of voters out of the pool of voters who didn’t vote in primaries. This time, it’s possible they could draw a disproportionately Democratic group out of the Republican-leaning pool of those who didn’t vote in the midterms.
Imagine, for instance, that the infrequent Black or young voters who say they back Mr. Trump in the polls generally don’t show up, while those who back Ms. Harris really do come to the polls. This does not strike me as especially far-fetched (and I wrote that sentence before a comedian at a Trump rally referred to Puerto Rico as an “island of garbage” and set off a torrent of social media criticism).
The Times/Siena polling suggests something like this is a real possibility. The poll asks voters how likely they are to vote, and Ms. Harris usually fares best among those who say they’re almost certain to vote, while Mr. Trump fares better among those who aren’t sure.
This should not be especially surprising, as Ms. Harris fares best among higher-turnout voters. What is more surprising is that this advantage penetrates even within different turnout groups. For instance, Mr. Trump leads by seven points among 2022 nonvoters who say they’re almost certain or very likely to vote, while he leads by 14 points among those who are less likely.
All of this information is incorporated into the Times/Siena poll, which estimates the likelihood people will vote by blending their track record of voting with whether they say they will vote in the coming election. Historically, both measures have independent value in predicting turnout. In this election, they point the same way: As a result, Ms. Harris has usually fared better among likely voters than registered voters in Times/Siena polling. We consistently find her ahead in Michigan, for instance, even as she consistently trails among registered voters.
This basic approach has validated well after elections: The electorate in the Times/Siena poll usually closely resembles the actual electorate, based on the final turnout data available months after the election. In fact, this is probably the one thing that hasn’t contributed to survey error in Times/Siena polling in recent cycles.
Nonetheless, this record does not necessarily mean anyone should be confident in this exact view of the electorate. In the past, Times/Siena polling hasn’t found such a stark relationship between these turnout measures and vote choice. As a result, it has been far less sensitive to these kind of choices — and the eventual turnout.
This cycle, it’s easy to see how this ends up going differently, with big consequences for the outcome and our understanding of the election. These scenarios can even be mixed and matched: What if white working-class people with no record of voting turn out for Mr. Trump, while the dissenting and disaffected young and Black voters stay home? In this scenario, many of the big demographic swings toward Mr. Trump that we’ve reported about this cycle might fade, but Mr. Trump could win nonetheless. There are countless other possibilities.
There’s another reason not to be so confident in how low-turnout voters will behave this cycle: They are also fairly likely to be undecided. Even during the last Times/Siena polls, they were far less likely to know whom they supported. As a consequence, they may have also been far less likely to know whether they would vote at all. As they decide, they may also decide to turn out — or not.
If these voters break one way or another as they tune in to the race, it’s easy to imagine how either side wins comfortably.
The post This Time, It May Actually Come Down to Turnout appeared first on New York Times.
