Bayesian Updating
5 min read · 4 principles · apply it by lunch
Good morning. You are allowed to be forty per cent sure. Almost nobody takes the option.
Name a belief you have carried about your business for a year — a conversion rate, a renewal assumption, a channel’s payback — and say what has arrived since that should have moved it a few points either way. If nothing has moved it, the reason is rarely that the evidence went quiet; it is that the belief never had a number on it, and a belief without a number has nothing to move.
Thomas Bayes was an English Presbyterian minister and amateur mathematician who died in 1761 leaving an unpublished essay behind him. His friend Richard Price edited it and sent it to the Royal Society, where it was read in 1763 as An Essay towards Solving a Problem in the Doctrine of Chances, and it became the standard grammar of evidence: a belief is a probability rather than a verdict, and evidence should revise it in proportion to how much better it fits one answer than the other. You hold a prior, evidence arrives, you land on a posterior — which is simply the prior for whatever comes next. Philip Tetlock, professor at the University of Pennsylvania, found the working version in the forecasting tournaments he and Dan Gardner describe in Superforecasting (2015): the most accurate forecasters were not the boldest, but the ones who updated often, in small increments, on ordinary evidence. Most operators do the opposite. They hold a position at full strength until something forces a reversal, then reverse completely, and call both halves judgement.
Bayesian Updating is Day 84 of 365.
In the Daily, each of its four principles is explained in full, with a one-page template to apply it the same day. A new framework arrives every morning at 7am for a year, across all fifteen courses of the FutureMBA curriculum.
$29 a month · cancel any time · 7-day money-back guarantee