Pi-Consulting

Turn on your TV and tune to any news service. Within three minutes on average, an “expert” will be telling you what will happen next on this or that big issue of the day. The news and information media industry has gotten so big that it can no longer subsist on ‘what already happened’ and ‘what is happening now’ alone. It requires a steady output of ‘what will happen in the future’. What started as the fanciful musings of Jules Verne and H.G. Wells has become the serious Business of Forecasting.

So how good are forecasters at forecasting? The writer and self-styled intellectual historian Louis Menand, writing in the New Yorker magazine, has answers.

How good? Sorry, not very. Menand reports that in a 20-year study conducted by Philip Tetlock, a Berkeley psychologist and academic, a group of experts on social/political issues were asked to make forecasts based in each case on three possible measurable results. They were scored on probability assessment and on the accurate prediction of outcomes. Collectively, these experts performed worse by applying their expertise than if they had randomly assigned equal probability to each of the three possible outcomes. As Menand put it, “Human beings who spend their lives studying the state of the world are poorer forecasters than dart-throwing monkeys”.

Why? Experts are too often fall prey to personal hunches, and proceed to fall in love with them, tending to dismiss new information that doesn’t fit their original pet theory. Who’s to contradict them, anyway? They’re the experts! Most of us discount new data that contradicts what we think, and “would rather find more reasons for believing what we believe than look for reasons why we might be wrong”. But then Most Of Us don’t pretend to be right all the time. Unlike the self-styled experts.

There’s more. Experts, gurus and pundits do so like to exhibit complex reasoning that make them look smart, whereas in the real world the odds tend to be with the obvious. If it walks like a duck and quacks like a duck, could it possibly be… a duck? Not when the experts get hold of it. They will find ways of persuading themselves that “The Duck-Like Object” is something vastly more intricate, mysterious and significant. Then they will try to persuade you of the same.

This is partly because of the irresistible human impulse to make things more complicated and portentous than they really have to be. And the more famous the forecaster, observes Menand, the more overblown the forecast. The pundits seem to feel that the more intricate the prediction, the greater the kudos associated with making it. Accordingly they profess to have discovered “a set of interlocking causes that no-one else has spotted”. This is rather foolish, since the likelihood of three separate things all happening at the same time is statistically much lower than the likelihood of any single one of them coming true. So the prediction is that “a stranger will come into your life”, huh? Fair bet. But “a TALL, DARK, HANDSOME stranger?”. Don’t hold your breath.

What happens when confidently-predicted outcomes prove to be wildly mistaken? If there is one thing experts hate more than being wrong, it’s being proved to be wrong. But they need not worry unduly. Much prediction is of the long-shot variety; by the time the answer is known almost everyone has forgotten the question. If confronted with getting it wrong, dear expert, tell ’em you were right on substance, just wrong on the timing. Tell ‘em a completely unforecastable random ‘act-of-God-type’ event knocked your forecast sideways. If all else fails, bluster; tell ‘em you were right all along, despite appearances. (Governments do it all the time, and seem to get away with it. Accountability? What’s that?).

Who knows what will happen next? Actually, YOU do, at least as reliably as all those experts you’ve been slavishly relying on. Thanks to the news media (the reporting, not the forecasts!), the internet, and factual resources like Pi Consulting, you have access to the same empirical facts on any issue as the gurus, pundits and hot-air-blowers. Make up your own mind. You’re just as likely to get it right as they are.

If you like, we’ll help.

(Source: Louis Menand, writing in the New Yorker; Pi Market Research)