Decryption | Biden, Trump and the polls

(New York) 231 days before a Biden-Trump return match made official during the primaries and caucuses held last Tuesday, the feeling of déjà vu does not extend to the polls. Because if the 2024 presidential election took place this Monday, the Democratic candidate, now president, would probably bow to his Republican predecessor, due in particular to the voting intentions expressed by voters in key states.

The reaction of Joe Biden and his camp to this reality is almost Trumpian. In a portrait dedicated by the New Yorker at 46^e president, one of his close advisers says that the polling techniques “are screwed”. “We can’t reach anyone on the phone,” he confided. Last Monday, during a trip to New Hampshire, his boss made the same criticism, specifying that we have to “make six zillion calls to reach a person on their cell phone.”

So what does a seasoned pollster think of these statements considered as denial by certain political analysts?

“It’s an attitude of losers,” responds Quebecer Jean-Marc Léger, whose company has been feeling the political pulse of Americans since 2016. “If the polls placed them five points ahead, they would not have the same reaction . They would let themselves drift on the polls. »

“Take Donald Trump,” he adds. He only publishes favorable polls on Truth Social. »

In fact, the Republican candidate is torn these days between his old habit of denigrating the polls and the desire to celebrate the most recent ones, as he did last March 9 during a speech in Rome, Georgia , more or less coherently.

“I think I’m doing well – you know, the polls are all rigged. Of course, these days they’re not rigged because I win by a lot, so I don’t want to say that. Disregard this statement. I really like surveys. »

In defense of Joe Biden and his camp, Jean-Marc Léger recognizes that the polls are not all reliable.

Some, carried out by pro-Republican or pro-Democratic organizations, produce results that are much more propaganda than science. Others use the least expensive techniques, including automated calls, which young people tend not to answer.

But there are quality pollsters, according to Mr. Léger. In addition to his own business, the Quebecer mentions duos New York Times/Siena College, CNN/SSRS and ABC News/Washington Post, as does Fox News. “As surprising as it may seem, Fox News produces relatively good polls,” he says.

A “reliable” technique

All these pollsters give the advantage to Donald Trump over Joe Biden nationally or in key states. But, more than seven months before the election, they are anything but predictive, including the one that Léger carried out in the United States at the end of February. This barometer credited the former president with 46% of voting intentions compared to 42% for the president.

“We’re measuring jello right now. An election is something that changes,” says Jean-Marc Léger. And the pollster added: “The polling technique is reliable. It is the voter who is no longer a voter. »

The phenomenon is perhaps not new. American history provides several examples of spectacular swings in voting intentions. At the same time in the 1980 presidential campaign, Democratic President Jimmy Carter was ahead of Republican Ronald Reagan by 14 percentage points. He lost the presidential election by 10 percentage points, a difference of 24 percentage points.

Since 1944, polls taken starting in mid-March have missed the final result of the national presidential popular vote by about 8 percentage points on average, according to data analyzed by the website 538.

That said, in 2020 the difference between the national polling average in mid-March and the final result was less than 1%.

But pollsters underestimated Donald Trump’s support in key states, as they did in 2016. According to Jean-Marc Léger, their desire to avoid the same mistake in 2024 is perhaps pushing them to make the reverse error.

“Look at the Republican primaries, pollsters everywhere overestimated support for Donald Trump,” he said.

“Magical” moments

In the long electoral marathon which will lead American candidates and voters to November 5, the date of the presidential election, Jean-Marc Léger sees three “magical” moments which are likely to have a significant impact on public opinion. These are moments “when the voter is psychologically open to the opponent’s arguments,” according to the pollster.

The latter first mentions the immediate period after the conventions of the two parties.

“There are definitely movements in the polls, more or less, after these events,” he said.

The first presidential debate is another one of those magical moments, he says.

But “the important thing is not the debate, it is the debate about the debate that matters,” he maintains. Kind of like what happened after the State of the Union address. The important thing was less the speech than the debate which followed and which was very favorable to Biden. Ultimately, this man is not senile or on the verge of death. »

Last magical moment of the presidential campaign: the very last week. In 2020, for example, Donald Trump succeeded, in the home stretch, in tightening the race in key states by increasing rallies and attacks against Joe Biden.

“And I am adding a fourth moment in 2024,” adds Mr. Léger. If we ever see Donald Trump in handcuffs and an orange suit, it’s over. »

At least, according to current polls.