If President Trump combined a winning electoral college on Tuesday, at least one pollster – and perhaps just one – would be able to say, “I told you so.”
That person was Robert Cahaly, with whom the Trafalgar Group this year released a consistent stream of battlefield-state polls that showed the president was highly competitive with Joseph R. Biden Jr., and often surpassed. to the lead, in states where most of the other prospectors showed a stable lead Biden.
Trafalgar did not disclose his method and was deemed too glossy by other polls to take it seriously. Mostly, they remove it as an exception. But for Mr. Cahaly, “I told you so”; was a business card.
In 2016, making its polls publicly available for the first time, Trafalgar was the company whose state polls showed Trump’s frustrating victory. A veteran Republican strategist, Cahaly even called the exact number of Electoral College votes Trump and Hillary Clinton would receive – 306 to 227 – despite his predictions of which states would Helping them get there is just a bit wrong.
So with free-to-flare concerns about whether to trust the polls, Mr. Cahaly dressed in ruffled, goat-like clothes and bow ties was recently asked on television news. cable. In addition to appearing on Fox News regularly, Mr. Cahaly took to CNN last week, explaining to Michael Smerconish why he thought the president would leave with an easy victory – and defending himself against a series of the criticism Mr. Smerconish made, one by one, from his colleagues Cahaly.
He said: Amid the pre-election media were searching for his theory of the case – it drove more than 1.5 million clicks to Trafalgar’s website on Monday, he said – question The big thing seems to be: Is it possible to believe someone whose polls keep coming up Mr. Trump is just enough to support a narrow lead in most of the revolving countries, and who refuses to reveal many things? About how he got his data?
In his last few polls this election season, Mr. Cahaly found Mr Trump to have a two to three point advantage in North Carolina, Arizona, Michigan and Florida, and lead elsewhere. That puts him ahead of most of the major polls, whose surveys in those states often show Mr. Biden has the upper hand. Different from everything this year, it’s hard to miss the 2016 aftermath, when Trafalgar took up a similarly lonely position the night before November 8.
Above all, Mr. Cahaly’s approach focuses on the belief that all people lie, but especially conservatives. This has been largely rejected by social science, but it has not diminished his beliefs. To hear him explain, the traditional polls (he called them “dinosaurs”) were paralyzed by “Social desire bias”: the tendency of respondents to say what they think interviewers want to hear, not what they actually believe. In Trump’s US, he said, that problem has gotten worse.
“I just think people don’t like what they say,” Mr. Cahaly said in a recent phone interview from Atlanta, where he lives. “We cannot eliminate desirable prejudice, we can only minimize it.”
Four years ago, he solved this problem by asking people if they would support the president and who they think neighbor will assist. This year, he said, he is using other means to achieve similar results.
But he didn’t say what they were. Mr. Cahaly gave almost no real explanation for his polling method; The methods page on Trafalgar’s website contains content that looks like a vague advertisement about their services and explains that their polls actively face socially desired biases, without Give specific details about how. He says he uses a mix of text messages, email, and phone calls – an automated and a caller-direct number – to achieve an exact representation of voters.
Conventional pollers, who adhere to a widely tested and effective method of collecting a representative sample, do not buy it. Besides, if there ever was something like a “shy Trump supporter” – someone reluctantly admitted that they intended to vote for the president – then that species would have almost gone extinct during the presidency Trump’s going on, said Daniel Cox, a conservative biased biased towards the American Enterprise Institute, a public opinion poll and pollster.
“People don’t seem ashamed to support Trump,” said Mr Cox. Over the past four years, studies that seek to quantify the so-called “shy Trump” effect in surveys have often found little evidence to support that.
Late last month, FiveThirtyEight’s Nate Silver got their hands on cross tabs of the still-in-progress Trafalgar poll in Michigan. It shows that more than a quarter of Democrats and Republicans are expected to vote for the other party’s candidate, so far far away from most other polls Silver calls. are “crazy” numbers.
Mr. Cahaly, of course, is not conducive to experts’ skepticism. He does not seem to care about whether he is following the best practices of the American Public Opinion Research Association, the commercial organization that carries the standard or not, moreover Mr. Trump said he was whether America’s NATO allies have respect.
Among fellow polls, the key is that Mr. Cahaly lacks transparency about his methods.
Josh Pasek, professor of communications, data and political science at the University of Michigan, said that without knowing the methods the company uses to reach survey respondents, it is impossible to rely on the children. number.
“It would be inappropriate not to tell me, not only which mode do you use to draw your pattern but also how you specifically did it,” he said. His general rule: “If someone is not transparent, you can often assume they are crap.”
There’s something undeniably compelling about the story of a quintessential, vibrant Southern pollster who entered 2016 with a new and proven all-door approach The larger row was wrong. Born in Georgia and raised in the South Carolina suburbs by a banker and a teacher, Mr. Cahaly developed a political obsession as a child and studied that major at the University of South Carolina. He quickly came under the direction of probe Rod Shealy, an aide to Republican strategist Lee Atwater, and eventually founded his own company.
Named after a battle during the Napoleonic War when the British navy returned French and Spanish ships at sea, Trafalgar, which he operated alone, conducted surveys on behalf of clients from year 2006.
Most of Trafalgar’s polls are conducted for conservative clients and Republicans, although – by a different standard of traditional standards – it is not reliably disclosed when the The survey is covered by partisan benefits.
In 2010, Mr. Cahaly was arrested and brought to court for violating laws against using automatic telephones – known as robots – to conduct polls. The charges against him were eventually dismissed, and he later successfully sued a state law enforcement agency, making South Carolina’s ban on robot calls declared unconstitutional. .
Mr. Cahaly said he was conducting a legitimate vote, aimed at truly understanding voters’ opinion – and getting what he called a “deadlock” result. During the 2016 Republican primary election, he soon noticed a surge of enthusiasm from many working-class voters, who had long felt alienated from politics and helped Trump go. up to power.
“I kept getting these stories about people who came to vote and didn’t know how to use the voting machine, they didn’t vote for too long,” said Cahaly. So he started looking at who those people could be and used the data available online to create a list of about 50 lifestyle traits – whether they own a fishing license or do they. no – to identify types of voters with low engagement. it turned out in batches. He used that data to ensure that he was reaching the right kind of respondent when he polled about voter records. before the general election.
In 2018, Mr. Cahaly once again achieved a successful track record in probing the Senate and governors races, including surveys that correctly identified Ron DeSantis and Rick Scott’s victories at Florida.
This year, he continued to see strong support from Trump among these voters and he believes other polls are underestimating their importance. Among Cahaly’s theories is that getting a conservative voter to complete a poll takes five times more than it would take to get a free poll. Others in the field say they found no evidence to support this in their own work.
But Mr. Cahaly asserted that poll participants will be confident that they are taking a representative voter sample just because they follow the scientific method. He went back to the political divide of the country, and how Americans today do not want to communicate with each other without doubt. In a sense, he has positioned himself as an advocate of Trumpism, giving a voice to a silent majority – or at least, the majority in the Electoral College – knowing that the elites consider their views. I am to blame, and therefore will not show them off freely. just anyone.
“Lee Atwater told everyone around me that you have to get out of the politicians’ head and become the head of Joe Six-Pack,” said Mr. Cahaly. “What do ordinary people think? And to do that, I want to talk to ordinary people. I like to follow up on poll calls and chat with people for 30 minutes.
Mr. Cahaly feels there is no need to reveal his technique, despite the near-widespread suspicion of his work from his colleagues. “I have given away enough; I won’t give it away, ”he said, arguing it would be a mistake to tell the public about his“ neighbor question ”, which a number of other companies have applied in their own surveys.
“I think we’ve developed something very different from what other people do and I’m not really interested in telling people how we do it,” he said. “Just judge us to see if we’re doing it right.”