scorecardresearch Skip to main content

Your guess is as good as theirs

The fall of the Berlin Wall, in November 1989, was a sign of the waning power of the Soviet Union, but few foresaw its collapse, in 1991.FRANCOISE CHAPTAL/AFP/Getty Images

Wondering what sort of winter Americans are in for, The Wall Street Journal last week checked the 2019 edition of the venerable Old Farmer’s Almanac, which is published each year in Dublin, N.H. It predicts a mild winter with “above-normal temperatures almost everywhere” this season.

To confirm that forecast, the Journal also checked the equally venerable Farmers’ Almanac based in Lewiston, Maine. It warns that Americans should expect “teeth-chattering cold, plentiful snow” and a “chilly, wintry mix.”

They can’t both be right. Yet each almanac is confident in its predictions, and each claims a high rate of accuracy. Does that sound familiar? Of course it does. It sounds like the endless parade of experts, insiders, and pundits who routinely predict the future, and routinely get it wrong.


The other day Politico rounded up a handful of the least prescient political predictions of 2018. Among them: Huffington Post reporter Matt Fuller’s prophecy that Joe Crowley would be the next speaker of the House, Carnegie Endowment scholar David Rothkopf’s assurance that the US embassy would remain in Tel Aviv, and CNN journalist Frida Ghitis’s forecast that President Trump’s approval rating would crumble to just 25 percent. In fact, Crowley lost his seat in Congress to political neophyte Alexandria Ocasio-Cortez, the US embassy was relocated to Jerusalem, and Trump’s approval rating remains above 40 percent.

It’s not only weather and politics that prognosticators botch. Financial predictions are notoriously unreliable. A paper released this year by the International Monetary Fund documented the decades-long failure of economists to spot a looming recession until it was already well underway. “Treasury yields have been forecast to rise every year for the past decade,” noted The Wall Street Journal in November 2017, “yet they have gone down more often than not.”


From military predictions to technological predictions to sports predictions, when experts foretell the future, it’s usually safest to assume they’re wrong. And just as that’s true of short-term predictions, it’s also true of long-term predictions.

On Jan. 1, 1901, at the dawn of the 20th century, the editors of The Boston Globe surveyed the international scene and assured its readers that “the rise of Germany” posed no cause for concern, since “of all the nations she is probably the least corrupted by the lust of conquest.” In 1968, the Foreign Policy Association published “Toward the Year 2018,” an anthology of predictions by a dozen “eminent leaders” of where the world would be 50 years in the future. A few auguries they got indisputably right — several contributors foresaw a world in which computers and information technology would play a central role. But many of their predictions, from man-made hurricanes used as weapons to antigravity cars, were laughably off-base.

It isn’t being snarky or arch to stress how inept experts are at predicting the future. It’s being prudent. Because they are deeply knowledgeable in a particular field, experts are more prone than others to view the world through a too-narrow lens, assuming that the current trends they understand so well are indicators of what is to come. Their expertise reinforces their confidence in their own analysis, blinding them to contrary data or disconfirming evidence. That helps explain why those who knew the most about the Soviet Union, for example, didn’t foresee its collapse in 1991. Or why judicial experts, such as constitutional law professors and former Supreme Court clerks, aren’t very good at predicting how the high court will rule.


In the 1980s, political psychologist Philip Tetlock asked 284 political experts to make roughly 100 predictions of future events, and to assign a degree of probability to their forecasts. Two decades later, with the benefit of hindsight, Tetlock was able to analyze the accuracy of the experts’ predictions. What he discovered was that they hit the mark only slightly more often than if they had guessed at random. Nonexperts who keep up with current events by regularly reading the newspaper, concluded the New Yorker in a review of Tetlock’s study, “can guess what is likely to happen about as accurately as the specialists whom the papers quote.”

That’s worth keeping in mind as 2019 gets underway, and from every quarter self-assured sages and savants materialize to tell you what to expect. As you listen to their smart, persuasive, credible prophecies, just remember: Most of them, most of the time, will be wrong. (You can take my word for it. After all, I’m an expert.)

Jeff Jacoby can be reached at