causal

Maximize Samples And Variance For Convex Returns

Scientific and artistic breakthroughs follow convex, power-law returns, so what you want is to increase the number of samples and push toward more variance—yet many existing institutional mechanisms not only fail to increase variance but structurally attenuate it, which is especially bad precisely because it suppresses the positive catastrophes (White Swans) where the value concentrates.

causalpending

Speaker

Patrick Collison

Evidence Quote

“in these domains with these kind of convex returns where, or some power law returns... what you want to do is you want to increase the number of samples and you want to kind of push towards more variance... there are all these mechanisms that actually seem not only not designed to increase the variance, but in many ways they kind of structurally attenuate us.”

Source

Patrick Collison on Innovation and Scientific Progress— EconTalk
Created: 6/13/2026, 12:29:12 AM

Your notes

Loading notes...