Admit it, big predictions are rarely a good idea

The best antidote to our prediction addiction is to start holding wayward pundits to account. Perhaps it’s time we all started keeping score.

Protestor holding up a banner during a demonstration in the city centre of Frankfurt, Germany.

Just over seven decades ago, not long before he died, George Orwell achieved what all writers envy: a prophecy that came true. It’s embedded in a short essay he wrote for Tribune, several weeks after Japan’s surrender, published under the headline, You and the Atomic Bomb.

In it, Orwell sketches his vision of the post-war order; an altogether fraught and vertiginous picture, dominated by rival nuclear superpowers, whose annihilatory impulses are held in check by the constant threat of mutually assured destruction. This crude calculus, he reasoned, was best summed up in the phrase “a peace that is no peace”.

There’s probably no other writer in the English-speaking world whose words are so routinely described as “prophetic” in quite the same way as Orwell’s. On this occasion, however, he truly does deserve the designation writer-sage. Who else, writing in 1945, foresaw the power dynamics of the Cold War with such scintillating precision and clarity?

In many ways, Orwell was a freak. He never dies. But in the chaotic business of forecasting and prediction, he should be regarded as a maverick outlier, the exception that proves the rule – and the rule is that most experts and pundits make for terrible forecasters.

There’s nothing particularly novel in this observation; we’ve known for some time, perhaps centuries, that experts tend to get the big stuff wrong. And yet almost every part of national debate is drowning in overconfident forecasts. Think dud election calls, sweeping macroeconomic forecasts or puffed-up pandemic predictions that turn to twaddle.

Our desire to know and chart the future is at its strongest in politics. Take the US election, just over. Yes, the polls were out. But what of the cocksure pundits who joined the 11th-hour flurry towards the Democrats?

The hosts of The Rest Is Politics podcast, Alastair Campbell and Rory Stewart, were so convinced of a Harris victory that, days out from the election, Stewart rebuked commentators who even suggested the race would be a close run thing.

He’s a polished observer, articulate and perceptive, but what did he know that others didn’t? He bet the farm on a Trump defeat. In the end, he showed how political projection and plain old wishful thinking often masquerade as expert prediction in disguise.

About a decade ago, the social psychologist Philip Tetlock claimed most experts have no idea what they’re talking about when it comes to forecasting. Tetlock concluded that the average pundit is roughly as accurate as a “dart-throwing chimpanzee”.

The best antidote to our prediction addiction, he suggested, is to start holding wayward pundits to account. Perhaps it’s time we all started keeping score.