If It’s Worth Doing, It’s Worth Doing With Made-Up Statistics

A defence of quantifying uncertain things: even made-up numbers encode real information and, run through ordinary arithmetic, beat unaided intuition, which fails badly at scope sensitivity and partisan consistency. Framed as the everyday version of preferring a Bayesian calculation to a gut feeling, with the CFAR Bayes anecdote and the scope-insensitivity and hungry-judge findings. Memorable line: "It's easy to lie with statistics, but it's easier to lie without them."

Quality 68 · Strong. A tight, vivid, useful articulation of a core rationalist principle and a standout explanation, but it restates rather than extends existing ideas (Tetlock / Kahneman / LessWrong). Strong, not Excellent.

Claude’s paradigm shift 30 · Slight. The thesis - quantify, because numbers beat intuitions - was already standard in the rationalist / forecasting milieu by 2013; this is a memorable restatement, not a new paradigm. Slight.

Real-world impact 3 · Moderate. A much-cited, widely-quoted rationalist principle ('worth doing with made-up statistics') that became a standard pro-quantification touchstone tied to consequential estimation/EA practice; reach is mainly intellectual but the coinage is broadly adopted → RWI 3.