Here’s a great historical example of – “I guess you had to be there.” From today’s perspective, it’s hard to see why the presidential election of the year 1900 produced a record turnout of eligible voters of more than 73 percent. Not only was the winner, President William McKinley a less-than-riveting personality, but 1900 was also the second time he had beaten the Democratic candidate William Jennings Bryan, and the results the second time around were – by and large – no different from the first.
But this statistic, 73 percent of eligible voters, drove a spate of articles about the historically high voter turnout for this year’s presidential election. 2020’s 66.6 percent of eligible voters, this popular meme went, was the highest in 120 years, referencing McKinley-Bryant II. But the comparison is apples and oranges because 1900 was before women got the vote, so no women were counted as eligible voters, so what the statistic shows is that 73 percent of eligible American males cast a ballot.
Toss out the spurious historical comparison and the size of the 2020 vote is only more impressive. An almost perfect two-thirds of eligible voters adds up to about 157 million, more than ever before. Donald Trump’s 73.7 million votes is more than every other presidential candidate in American history, except, of course for the man who beat him. Joe Biden’s vote of 79.6 million is the present all-time record.
The size of the record turnout was one of the several things that most pollsters got wrong in their pre-election predictions. University of Florida professor Michael McDonald’s guess of 150 million was considered shooting high. Actually, it fell short by about seven million votes.
Of course, who won or lost, and by how much, are of more interest than how many played the game. And by these standards, most of the pollsters fell even farther short of reality. In a New York Times survey of name-brand national polls, all of them predicted an easy Biden victory, with a clear margin of eight points or more. The reality: Biden won, but more narrowly, by 3.8 percent…51 percent of the vote to 47.2 percent.
Here are two interesting comparisons. In 2016, pollsters also overestimated the Democrat Hillary Clinton’s popular vote margin over Donald Trump. The prediction, Clinton by four percent; the reality, Clinton by two. In 2020, the difference in votes between Trump and his Democratic rival doubled, both in prediction and reality. But the scale of the error – a prediction off by 50 percent — stayed exactly the same. As my Grandma would say – “Oi!”
I like that our guest again today John Zogby prognosticated more correctly than most of his rivals, and I especially like that he pointed out where his colleagues might be going wrong on this program over the course of the campaign, but what I love is his description, in a recent op-ed in the Miami Herald of how he does it. “This is artistry,” Zogby wrote, “rooted in long years spent enumerating, capturing the heart and soul, and defining the drivers of real voting patterns.”
And then he reveals his secret weapons beyond his experience and insight, he has eyes and ears. “It helps that I live in a region where Republican and Democratic lawn signs are common and as is talk among those who are passionate about both sides.”
John Zogby — pollster, author, trend-spotter, and thought leader — has spent the past four decades as one of the most accurate pollsters in the world, conducting business in 80 countries, and leading the way in finding the meaning, story, direction, and usefulness of the data collected.
His client list is a Who’s Who of Fortune 500 companies (GE, Microsoft, Cisco Systems, Coca-Cola, IBM); global NGO’s (UNAIDS, the World Health Organization); and government agencies (the US State Department, US Department of Defense, the Mayors of New York City, Houston, Miami).
John’s regular columns for Forbes.com and the Washington Examiner dissect the intersection between cultural values and political behavior.