Ben Recht's forecasting lecture argues probability conflates frequency and belief

beenwrekt · x · 2026-09-10

In Class 5 of his graduate seminar "Forecasting: A Critical Retrospective," Ben Recht argues that probability is unavoidable in forecasting but used far too casually. Using weather prediction as an example, he shows how data-assimilation outputs get interpreted as probability distributions, sampled into future scenarios, and then transmuted into beliefs about outcomes — a slippery shift between frequencies and beliefs with no well-specified formal rules. Drawing on Paul Meehl and Rudolf Carnap, he distinguishes logical probability from other notions and contends every applied probability course should explicitly call out this transmutation.

Original post →

More from AGI Musings

AGI Musings channel →