James Carville was on Manhattan’s Upper East Side recently when he was stopped by anxious New Yorkers begging a popular Cajun political sage for some kind of guidance on how the November election would turn out.
There was nothing for them.
“They strongly believed that I had some kind of secret knowledge,” the veteran Democratic strategist said. “The hardest thing in the world is to tell someone who thinks you can do anything that you’re not.”
The hard-to-swallow truth for anxious voters of all stripes is that no one can predict the future, especially in what seems like the closest election in most Americans’ lifetimes right now, when the outcome could be more decisive.
Weeks before Election Day, the race between Vice President Kamala Harris and former President Donald Trump is deadlocked, with polls showing razor-thin margins in key battlegrounds and nationally.
“The public and private in this election is closer than I’ve ever seen it,” said Dan Pfeiffer, a former aide to Barack Obama. “Everything is within the margin of error, so we’re at a point where the polls can’t tell us much about which way the horse race is going.”
A recent poll by NBC News According to Democrat Jeff Horwitt, who polled with Republican pollster Bill McInturff, the race is split exactly 48-48% — a “dead heat.”
The story is the same in other polls and polling averages, as Quinnipiac University pollster Tim Malloy said the race “couldn’t be any closer.”
Few analysts are willing to bet today after the vote was missed in the last election. There is zero expert consensus on the outcome, let alone the consensus view that has defined the final days of the last four presidential campaigns — in large part because much of that conventional wisdom has been proven wrong.
An entire industry has sprung up to analyze data and filter it through complex probabilistic mathematical formulas. But in a country that is balanced on the tip of a pencil and can fall in any direction with the proverbial wind, analysts are still looking for synonyms for the word “toss up” in all the data in the world.
“Now it’s literally 50/50” Nate Silver he wrote X Tuesday. Republican pollster Frank Luntz summed up the election with a single gif: Coin flip.
Republican strategist Matt Gorman looks back at Florida’s 2000 election by 537 votes and the extreme legal battle that went all the way to the Supreme Court.
“This will be the closest since the fight against Bush Protection,” Gorman said, noting the possibility of similar legal battles after the election. “There is more likelihood and political motivation for both sides to move beyond this issue.”
“Hope you’re ready to eat turkey in Erie County!” he joked about reporters and campaign workers as he descended on a battleground in northwest Pennsylvania to watch ballots be counted on Thanksgiving Day.
Certainly, almost every election in the 21st century has come close.
Swings like the 49-state sweep that led to Ronald Reagan’s re-election in 1984 seem foreign in a political era where Barack Obama’s 7 percent victory over John McCain in 2008 felt like a rout. Voting at polling stations was as close as it is now or even tighter 2004 and 2012 elections.
The rise of “big data,” both within the campaigns and for outside observers trying to understand them, helped create a sense of certainty that the 2008 results — if not the edges — were correctly predicted by the data, perhaps never quite right. and the 2012 elections.
After most pollsters and prognosticators missed his 2016 victory, the Trump era has brought the return of uncertainty, and 2024 is the narrowest of his three elections.
In 2016, all the smart money was on Trump losing, so his narrow victory over Hillary Clinton stunned observers and undermined confidence in the polls. In 2020, the data correctly predicted a Joe Biden victory, but Trump came much closer than many expected.
The opposite happened in the 2022 midterm elections, where the Democrats outperformed expectations and there was a well-established historical trend of midterm elections going against the party in power.
“What we actually have is an unusually stable electorate — because of partisan tribalism and cultural identity politics — that’s evenly split,” said Monmouth University pollster Patrick Murray. “So little movement in the electorate could easily swing the outcome either way.”
The Electoral College as a whole adds an extra layer of uncertainty because it can add up to small swings in individual states, giving the winner all the electoral votes, whether they get 1 vote or 1 million.
“The problem with experts is that they can’t deal with constant uncertainty,” Murray said.
In fact, the polls in 2012 trailed by a wider margin than in 2016, but people didn’t pay as much attention to it as they did in 2016 because the polls underestimated support for an Obama victory.
So while the polls underestimated Trump’s support in 2016 and 2020, they underestimate Harris’s support, just as they did for Obama today.
And the vast majority of the electorate seems almost impervious to new information, as the controversies, bluffs, assassination attempts, and hundreds of millions of dollars in campaign ads seem to have no effect on the election or simply cancel each other out.
“This race is notable for both its closeness and the insane consistency of the polls in recent weeks,” said Republican pollster Kristen Soltis Anderson. “This race just doesn’t budge.”
But as tight as the race is right now, none of that necessarily means the final result will look like today’s vote. With the electorate rushing in both directions and the Electoral College’s empowering influence, anything from a decisive Harris victory to a big Trump victory is possible.
“I believe the vote will be understood now,” Carville said, “but I don’t think it’s going to end that way.”
He and some others — but not all — think the seven battleground states are more likely to break mostly in one direction or the other: “For me, the most difficult scenario is breaking 4-3.”
But that answer doesn’t satisfy the worried Manhattan liberals he met the other day. So he suggests something else: “Democrats haven’t lost an election since the summer of 2022.”
“I don’t know what that means,” he said. “But it has to mean something.”