Before President Joe Biden abandoned his reelection bid in July, there was a broad consensus about where the 2024 race for the White House was headed. Former President Donald Trump was going to win and Biden was about to lose, with the only remaining questions involving the margin of victory and how many downballot Democrats the incumbent was going to drag down in defeat with him.

Even then, there were still occasional reputable public polls that showed Biden with a fighting chance. An NPR-Marist survey released after the fateful June 27 debate actually showed the soon-to-be defenestrated Democrat leading by 2 percentage points nationally. That was an outlier, but Fox News and NBC News polls that came out around the same time gave Trump leads of 1 and 2 points, respectively, well within the margin of error. 

Other polls, including Democratic internals, showed Biden not only losing but jeopardizing his party’s hold on Virginia, New Hampshire, Minnesota, New Mexico, Colorado, and New Jersey while losing the popular vote nationally. That is a bigger collapse of the blue wall that Hillary Clinton experienced in 2016, when Trump was elected the first time. Biden still maintains he could have won, most recently in an appearance on The View, but few Democrats believed that. But he bowed to pressure from his own party and dropped out weeks before the Democratic National Convention.

Since Vice President Kamala Harris replaced Biden at the top of the ticket, Democrats have felt much better about their chances in November. But the consensus about the state of the race has totally collapsed. Wags have dubbed this the “choose your own adventure” election.

Some polls show Harris not only outperforming Biden’s ill-fated reelection campaign but his successful 2020 performance. CBS News and the Democratic Data for Progress have Harris leading by 4 points, on par with Biden’s final popular vote margin four years ago. NBC and Morning Consult each have Harris up by 5, with Reuters-Ipsos giving her a 6-point advantage. By comparison, Barack Obama beat John McCain by 7.2 points nationally in 2008, the most lopsided presidential contest of the past 24 years.

At the same time, many high-quality polls are still showing a much closer race. CNN has Harris leading by just 1 point. New York Times-Siena College and Quinnipiac have the race tied, with the former also showing Trump maintaining his lead in the Sun Belt states where Harris surged after taking the baton. A few, such as Rasmussen, even still have Trump leading, however narrowly, nationwide.

Gallup is projecting a favorable national climate for Republicans. “The political environment suggests the election is Trump’s and Republicans’ to lose,” wrote the venerable polling organization’s Jeffrey Jones. “Nearly every indicator of the election context is favorable to the Republican Party, and those that aren’t are essentially tied rather than showing a Democratic advantage.”

It wouldn’t be 2024 without caveats, however. “Nevertheless, the two major party presidential candidates have similar favorable ratings in Gallup’s September poll, echoing presidential preference polls that suggest a neck-and-neck race between Trump and Harris,” Jones added. 

“If Trump were to lose, it suggests that concerns about his style, character, temperament, felony charges and convictions, and age outweigh Americans’ anxieties about the state of the nation,” he concluded.

Pluralities in Arizona, Georgia, and North Carolina said Trump’s policies helped them personally, according to the New York Times-Siena College poll. Smaller pluralities in each state said the opposite about Harris’s policies. A CBS News-YouGov poll came to similar conclusions in Georgia. Nearly half said they would be “financially better off” under Trump’s policies compared to 35% who said the same about Harris’s, while the “financially worse off” numbers were 45% for Harris to 34% for Trump.

The national RealClearPolitics polling average at this writing has Harris leading by 2.1 points, which wasn’t good enough for Clinton on Election Day in 2016 and is within the margin of error. That’s an improvement over Biden, who trailed in the same average by 3.1 points at the end of his campaign. But with Trump clinging to the slimmest of leads in the top battleground states, it still looks like a coin flip. 

“This race is TIGHT: within 2 points in all 7 critical battlegrounds,” CNN election forecaster Harry Enten wrote on X. “Harris barely wins if the result matches the polls.”

That’s all assuming the polls are roughly accurate. Trump has outperformed his poll numbers in each of the last two presidential elections. Democrats, to some extent, outperformed theirs in the midterm elections, especially with regard to a national environment that looked tilted toward the Republicans but thanks to the increased salience of abortion after Roe v. Wade was overturned that year (and anti-Trump sentiment) was actually more even. 

Harris would not be able to withstand a polling error in Trump’s favor based on her current numbers. She could romp to the finish line if the polls are somehow underestimating her. Republican strategist Ryan Girdusky is among those who detects liberal response-rate bias in Harris’s favor in a number of polls testing her against Trump head-to-head. 

“So how does a poll go from being statistically tied to a Harris landslide? She improves with white voters while mitigating Trump’s improvements with men, blacks, and Hispanics,” he wrote in his newsletter. “If Trump either increases with men or doesn’t lose as much with white voters, it all falls apart.”

Multiple polls show Trump running up big margins with voters who skipped the midterm elections and the 2020 election. This is part of what is prompting the Trump campaign’s emphasis on turning out low-propensity voters, leading them to employ nontraditional surrogates such as Robert F. Kennedy Jr. and former Hawaii Rep. Tulsi Gabbard. 

It is a risky strategy because low-propensity voters are so described for a reason — they don’t reliably show up to vote. Harris has inherited a more proven ground game from Biden, with major staffing advantages over Trump in the battlegrounds, while also raising more money.

Nevertheless, it is entirely possible that the election ends up being historically close like 1960 or 2000, except with multiple states reprising the role Florida played in the nail-biter between George W. Bush and Al Gore. “This could go on for days or even weeks,” a veteran Democratic strategist told the Washington Examiner. “The election lawyers will get their billable hours in.” There are also plausible scenarios in which either Harris or Trump sweep the battleground states and there is little suspense whatsoever. 

The uncertainty is already driving political observers mad. SFGate’s Drew Magary angrily denounced the New York Times for describing the race as “deadlocked” rather than tilting in Harris’s favor. “You don’t have to work terribly hard to sum up this race as it stands: Harris is destroying Trump, because Trump is a deranged old s***bag. See how easy that was?” he wrote. Press Watch’s Dan Froomkin called it a “must-read column” exposing “how the New York Times is boosting Trump’s candidacy by refusing to report the truth: That Harris is destroying him.” 

This isn’t exactly a universal view even on the Left. “At this point in the race in 2016, Clinton had a big polling lead over Trump in those states. But Trump voters were ‘underestimated in the polls,’ and Trump ended up winning both Michigan and Wisconsin — and the presidency,” the progressive activist group MoveOn warned in a Sept. 17 fundraising appeal. “The bottom line is this: This Electoral College is biased in favor of Republicans, the polls are uncertain, and the only thing we know for sure is that the election is a toss-up and will be won or lost by a tiny margin of votes in a small number of battleground states.”

Fear and panic work better in fundraising than confidence and complacency. Few prospective Democratic donors will be more likely to open their wallets if Harris is “destroying Trump.” Partisan hyperbole aside, however, this is an entirely reasonable read on the race. But after months of worrying about Trump defeating Biden, plenty of Democrats would be surprised if Harris lost.

On the other end of the spectrum, many grassroots Republicans are primed for Stop the Steal 2.0. Trump himself was willing to risk a preelection government shutdown over an attempted crackdown on illegal immigrants voting that Democrats viewed as a poison pill in a government funding measure. Some Republican operatives also worry his 2024 campaign is devoting too many resources to preemptive strikes against voter fraud relative to turnout and mobilization. 

Trump succeeded in persuading many rank-and-file Republicans that the 2020 presidential election was rigged and stolen from him. That was in a year in which he trailed throughout the campaign in the overwhelming majority of polls. If he loses this time, it would come after numerous polling leads dating back to 2023, multiple indictments by Democratic prosecutors, multiple failed assassination attempts, efforts to remove him from state ballots on 14th Amendment grounds, and the replacement of the Democratic nominee he had been favored to beat after the primary process had already concluded. 

It was the most extreme example, but it wasn’t the first time in the past quarter of a century that a losing candidate alleged foul play and their supporters believed them. Clinton conceded in 2016 and attended Trump’s inauguration. Gore did both of these things and even presided over the certification of Bush’s Electoral College win in one of his last acts as vice president. Neither of them ever stopped casting doubt on the legitimacy of those elections in other ways.

Democratic partisans echoed the phrase that Bush was “selected, not elected,” after the Supreme Court ruled in his favor in Bush v. Gore. Years later, Democrats told pollsters they thought Russia altered the vote tallies in 2016. They did not storm the Capitol as a result or embrace cockamamie constitutional theories about how they could overturn the results. The main proponent of the theory that John Kerry really won Ohio, and thus the presidency, in 2004 was, ironically enough, current Trump surrogate RFK Jr. rather than any mainstream Democrat.

But partisans of both stripes will be receptive to conspiracy theories if the election outcome remains an open question for too long. Any contemporary equivalent of the hanging chads and butterfly ballots chaos of 2000, alleged voting machine mishaps in 2004, the Russian influence campaign of 2016, and the COVID-19 voting protocols in 2020 will feed skepticism on the losing side.

Election officials and media outlets are bracing for uncertainty, slow vote-counting, and delays in declaring the winner. A Republican city commissioner in Philadelphia told CNN the odds of finding out the next president that night was “almost zero,” while the executive director of the North Carolina State Board of Elections predicted, “It’s going to be a late night.” The Boston Globe recently ran a story headlined, “Here’s why you shouldn’t expect to know the results of the presidential race on election night.”

The interim registrar of voters in a major Nevada swing county has temporarily taken leave from her job to deal with the stress stemming from a likely close election. The official “has requested a leave of absence for self-care,” a Washoe County spokesman told the Nevada Independent, affirming a commitment to both her “health and well-being” and “running a smooth and fair election.”

It took most news organizations at least four days to call the last presidential election. Republican voters felt they went to bed with Trump ahead in the battleground state vote counts only to wake up to those leads eroding or vanishing completely. Trump pushed their electoral doubts as far as he possibly could, and then further still to disastrous results on Jan. 6. But major cities boarded up their downtowns ahead of Election Day in anticipation of what might have happened if Trump had pulled it out instead.

A recent CBS News poll found that 58% believe there will be a greater risk of post-election violence if Trump wins compared to 42% who think the risk is higher if Harris prevails. That doesn’t appear to reflect fear of an anti-Trump backlash: 83% of Harris voters fear a worse threat of political violence following a Trump win, while 63% of Trump voters express the same concerns about the aftermath of a Harris victory. 

There is still the possibility that one candidate will win by a big enough margin that it will be difficult for many people to blame the result on voter fraud or suppression. In a time of intense political polarization, slow vote-counting, and no clear-cut front-runner, many will expect the worst until they see otherwise. 

W. James Antle III is executive editor of the Washington Examiner magazine.