Can you trust surveys? The fever of predictions takes over the 2024 elections

Who is going to win the 2024 presidential race?

That question irritates the country, since all kinds of forecasters, whether pollsters, academic or international probability makers — advertise your data and intuition to voters eager for a glimpse into a crystal ball that predicts a future Kamala Harris either donald trump administration.

No one knows what will happen on November 5th, since tens of millions of Americans have already done so. they already cast their ballots as part of early voting and with tens of millions more still to vote their preference on who should lead the nation for the next four years.

But the country is anxious about the outcome amid an increasingly bitter political divide.

More than 7 in 10 American adults say the future of the nation, its economy and its politics are a “major source of stress” personally, according to a study. report published this month by the American Psychological Association.

“It shouldn’t be passive. It shouldn’t be something that people don’t speculate about because it’s really important and this particular election is critical,” said Imani Cherry, a professor of media and public affairs at George Washington University. .

“There are a number of very, very, very important issues that concern millions and millions of people,” Cherry added.

As the unprecedented election between Harris and Trump draws to a close, experts say it is irresponsible for political observers to predict a winner.

“Polls are not predictive. It’s a snapshot in time,” said Republican pollster Robert Blizzard. “My job is not to try to predict an outcome, my job is to use surveys to help my candidate’s or client’s cause or problem succeed.”

The 50-50 race has been largely unmoved by major news events, including the president Joe Biden’s departure in summerpersistent economic problems, Harris’ historic bet and two assassination attempts against Trump.

Those who spoke to USA TODAY said any prediction of Trump or Harris winning the White House will be fraught with flaws thanks to the avalanche of polls in recent days, along with partisan polls aimed at swaying the base, the sports bettors who They seek to make money. and prediction models using companies with dubious track records.

Others point out that there has been a notable lack of quality polls in swing states that typically appear in the final weekends of a general election, while some say it might be time for newsrooms to reconsider relying on stories. of the surveys.

“People need to get off the rollercoaster of polls. They need to take a step back because people are playing games and playing games in our faces,” said Democratic pollster Cornell Belcher, who worked on Barack Obama’s two presidential campaigns in 2008 and 2012.

“It is not the task of a survey or of pollsters to predict the future, because we cannot,” he added.

Predict the winner of the horse race

The United States has not had a double-digit White House race since 1984, when incumbent Republican President Ronald Reagan cruised to victory by winning 49 of 50 states against Democratic challenger Walter Mondale.

Elections have been closer since then, including in 2000 and 2016, in which the Democrat won the popular vote but fell short in the Electoral College. Added to this is a hyper-competitive media landscape and now, in 2024, a cottage industry of pollsters, pundits and even gamblers seeking to give each side – and nervous undecided voters – a preview of what will happen.

Chief among them is FiveThirtyEight, which since 2008 has become the most popular database that strategists, news organizations and average viewers often cite in the race for the most powerful office on the planet. Uses a complex statistical model on potential outcomes and currently favors Trumpwhich wins 51 of 100 simulations.

But critics say people shouldn’t put too much stock in these models, since they often use polling companies with different accuracy ratings.

FiveThirtyEight, for example, reminds its audience that Harris once had a better chance of winning weeks ago and that could change if “some good polls” appear in favor of the incumbent vice president.

The site notes that when the odds of winning are around 50, it’s “a little better than a coin flip for the frontrunner,” and its founder, Nate Silver, wrote an op-ed in the New York Times that said: “50-50 is the only responsible forecast” during this cycle.

And the models are not always correct, as in the 2012 election, when then-President Obama defied the trajectory to defeat his Republican rival Mitt Romney. Similarly, Trump beat forecasts predicting Democratic candidate Hillary Clinton would win in 2016. Many companies also underestimated his performance in 2020.

What worries experts in the home stretch of the 2024 race is how predictive models could be influenced by partisan efforts to boost a candidate’s polling averages, which can be risky in the age of misinformation. as packs of supporters are likely to assume their candidate should win based on polling trends or early voting data.

The New Republic reported earlier this month on An avalanche of polls aligned with Trump is published during the summer with the aim of influencing the change of electoral forecasts in their favor.

For Trump supporters, especially, there is concern that his loss could spark a backlash similar to that of the Jan. 6, 2021, Capitol riot should the former president lose.

Blizzard, the Republican pollster, said it hasn’t conducted any polls for the Trump campaign this year, but it’s sure there are companies on both sides of the partisan divide that only highlight data showing their side doing well while They bury bad polls.

“If I don’t know the pollster, meaning they don’t have real clients, a clear track record, or only do horse racing polls, I’m less likely to care about their numbers,” he said.

Most campaign polls have ended at this point in the race, Blizzard said, and legitimate companies have moved on to using their data to make decisions about advertising and other resource allocations.

What experts find more troubling than bad actors seeking to manipulate the narrative is that few people understand that a point or two lead is a statistical tie that can go either way.

Cheery, a professor of media and public affairs, said the press plays an important role in this conversation because polling stories are easier to convey to the public.

He said more emphasis is needed on the consequences of Harris and Trump’s political differences.

Instead of making predictions, Cheery said he focuses his students on grassroots conversations among voters about what’s at stake in the race for the White House and ways people can participate in the democratic process through a entertainment-style coverage of who will win.

“It’s not that I don’t think polls have a place in this political discourse, but they are often very short-sighted,” he said. “Especially when we’re really talking about numerous variables that are going to come into play in people’s elections and for people who can say, ‘all the polls said (Trump) was going to win, and he didn’t; it was rigged.'” . .'”

‘A dubious and dangerous business’

About 34 million Americans have already voted early in 2024, and Democrats have a slight lead. according to the University of Florida Electoral Laboratorywhich tracks the numbers daily.

It shows that about 41% of votes cast come from registered Democrats, compared to 35% from registered Republicans, for example. But those who have measured the mindset of American voters for years say there is also a danger of delving too deeply into early voting data.

For starters, much of the data on early voting is based on the states where people register by party. Another caveat is that the baseline comparison is the 2020 election, when ballot access was expanded for the first time in response to the global COVID-19 pandemic.

Cathy J. Cohen, a political science professor at the University of Chicago, has been surveying black and Hispanic voters for months as part of GenForward, a youth project at the university.

The average American doesn’t realize the limitations or nuances of survey data, he said, including sample size, demographic representation and how survey questions are worded and ordered. They all impact the outcome of each survey, Cohen said.

“There’s a big difference and a big gap between people who take a survey and people who put on their coats, get in a car, take the bus, wait in line and make sure they’re registered to vote,” Cohen said. .

Instead of paying attention to possible outcomes, political observers should focus on trends over the course of an election cycle. But that hasn’t stopped other groups and teams from filling that void as more established firms appear to be taking a step back in 2024.

Many offshore betting markets, such as Polymarket, the world’s largest crypto trading platformhave been cited by Trump and his allies, with predictions that the Republican will win a second non-consecutive term as president. A French merchant He allegedly bet a total of $28 million in four different stories about the Republican candidate’s return to power.

Joshua Barton, a spokesman for BetOnline.ag, said betting on US elections on everything from who will win to turnout levels has skyrocketed in popularity over the past decade.

“In terms of the amount bet, it will eclipse what the Super Bowl does because there are a lot of high bettors who want to know who will win this election,” Barton said in an interview.

Some may never bet on anything else for another four years, he said, but participants want to act on the outcome.

It is not yet clear to what extent these and other criteria like the stock marketLevels from before the presidential election can predict who might prevail between Trump and Harris. But it worries Cohen and other academic experts who say surveys should focus more on identifying how Americans think, rather than telling the future or making money.

“They can be used to give some sense of prediction, but I would be cautious about putting too much weight on our idea of what’s going to happen based on polls,” she said.

“It’s a dubious and dangerous issue when it comes to human behavior.”