It is one of the legendary photos in the history of the United States elections. President Harry Truman waves a big smile at the front page of the Chicago Tribune with the headline “Dewey defeats Truman.” In its first edition on November 3, 1948, the newspaper announced the victory of the Republican candidate in the presidential elections, trusting the criteria of its Washington correspondent, who had been correct in four of the last five elections, and the polls that gave it for sure.
With 150,000 copies already printed, the news arrived that the scrutiny revealed a very close duel and they had to change the headline. A couple of days later, as Truman was returning victoriously by train to Washington, someone handed him a copy of the Tribune and the wrong headline with which he posed for the photo.
Something similar happened a few days ago with the legislative elections on Tuesday. The New York Times chose for its first edition a generic headline, but one that anticipated the victory of the Republicans: “The GOP (Republican Party) collects the first victories in key elections.” It wasn’t as dramatic a mistake as the one in the Chicago Tribune, but it was based on the same thing. The journalists were clear that Joe Biden’s party was going to suffer a clear defeat and they just had to wait to find out the difference.
A headline that is going to be corrected in subsequent editions of the newspaper and on the website is not the biggest mistake that can be made in the journalistic coverage of an election. The news and headlines chosen throughout the campaign have more influence. On October 19, The New York Times headlined: “The Red October feared by the Democrats has arrived” (red is the color of the republicans). “All the indicators on my political dashboard flash red,” wrote the author of the article, which was based on the results of a survey commissioned by the newspaper and on the fact that inflation always hurts the party in power.
A week later, the newspaper had the same idea (“reality is setting in”) after realizing that Republicans were pouring huge amounts of money into ads in states that are typically pro-Democratic.
It was not the only means that he had assumed that interpretation of events that had not yet occurred. ‘s main political correspondent was so clear two weeks before the vote that he already had a winner, one who was not running for election: Why the mid-term elections? They’re going to be great for Donald Trump.”
That headline didn’t age very well either. The general consensus in the media now says that Trump is the big loser, because the Republicans have not won the big victory they expected, which he could appropriate. In addition, the one who may be his great rival in the Republican primaries – the governor of Florida, Ron DeSantis – did achieve an unappealable victory.
Accustomed to looking for reasons that justify their own forecasts and polls, the media reported that Biden’s high inflation and low popularity, with a support level of 42%, would end up causing a clear defeat for the Democrats. As they had decided that the right to abortion would be one of his campaign prioritiesmany journalists decided in advance that this strategy had been a mistake, which turned out not to be true.
The final result of the elections is not yet closed. Two seats remain to be allocated in the Senate. If the Democrats win Nevada, they will retain control of the House. Otherwise, we will have to wait until December when the elections are repeated in Georgia, where no candidate exceeded 50% of the votes.
In the House of Representatives, the scrutiny has 21 seats pending. NBC News’ predictive model calculates that Republicans could reach 220 seatsonly two above the absolute majority.
Elections are always a unique challenge for journalists. Everyone wants to know who is going to win at the polls and there is a strong temptation to think oneself so intelligent as to know the outcome before others. The most common instrument is the surveys to which is added the personal impression of the political events that have taken place in previous years. A government or party is strong or weak based on an assessment that is always subjective, even though it may be reasoned based on certain facts.
The business of predictions is usually very risky. When you get it right, you don’t usually appeal to fortune, but to your own intelligence.
The commitment of the audiovisual media is more evident. In television talk shows, the presenters are constantly asking questions to find out what will happen in the future, especially if there are ballot boxes involved. Who will win the elections? with what difference? Journalists and political scientists put on the face of having reflected on the matter and offer their verdict.
The surveys lead the way. From them, a story is structured that explains why a match is ahead in that prediction or behind. A bit like the stock market. If it goes up or down significantly, it has to be due to something that happened in the past hours or days. Everything has a logic that is presented as self-evident, although no one is in a position to guarantee it completely.
In the US, the average of the polls this year has been closer to the final result than in the 2016 and 2020 presidential elections. Even so, many have been far from the verdict of the polls in several states, which is very important in such an even contest.
In theory, journalists can somewhat shield their forecasts by resorting to expert opinion. It is also true that you have to know how to choose them well and then not assume that all the experts think the same as those with whom you have spoken. You don’t have to be suspicious to suspect that people are sometimes sought out to confirm the journalist’s premises in an indirect version of the confirmation bias.
And then there’s the fact that experts are also human beings who make mistakes. Few more obvious examples than an October 2018 headline that offered a stark conclusion: “Experts believe that Vox should not worry the PP because in the extreme right ‘there are no votes’”.
Six months later, Vox obtained 2.6 million votes and 24 seats in the April 2019 elections. And in November, 3.6 million votes and 52 seats.
Although it may not seem like it because of how sure journalists are always, the bad thing about the future is that it hasn’t happened yet.
