The average expert was a horrific forecaster

The average expert forecast was dreadful. The television booking remained excellent.

Certainty is memorable, quotable and easy to arrange beneath a lower-third graphic. “There is a 63% chance, depending on these conditions” lacks the satisfying furniture of “This will never happen.” Unfortunately, events labelled nearly impossible kept happening and sure things kept refusing to arrive.

Expertise still matters. Replacing specialists with the loudest amateur merely diversifies the overconfidence. But knowledge of a subject and skill at forecasting its future are different products, however often sold in the same suit. Complex systems punish absolute language because the thing omitted may be exactly what changes next.

Ask for probability, time window and evidence that would move both. Keep score over dozens of predictions, including the boring ones television would never book. Reward updating before events make it compulsory. A product leader who moves from 80% to 55% after new customer evidence has not become weak; they have used the evidence. The colleague who stays magnificently certain has perhaps used a mirror.

Organisations claim to want accuracy and promote decisiveness. That bargain produces confident fiction. The future does not simplify itself because a senior person removed the qualifying words. It merely enjoys the surprise.

Behavioural principles

Behavioural ideas at play in this post

Short, plain-English explanations of the principles behind this post, with links to related books and examples in the archive.