a builder's codex
codex · operators · Morgan Housel · ins_housel-tail-events-drive-results

Tail events drive the majority of results — you can be wrong most of the time and still succeed if the few right calls are big enough

By Morgan Housel · Partner Collaborative Fund; author The Psychology of Money, Same as Ever · 2020-09-08 · book · The Psychology of Money — Tails Drive Everything

Tier A · TL;DR
Tail events drive the majority of results — you can be wrong most of the time and still succeed if the few right calls are big enough

Claim

A small fraction of outcomes — the tail events — drive the majority of results in most domains that matter. Venture investing, content creation, R&D, founder bets, and career-defining moves all share this structure. The frequency of correct decisions matters less than the magnitude of the few correct ones. You can be wrong the vast majority of the time and still succeed if the few right calls are big enough.

Mechanism

Power-law distributions concentrate returns in a small number of extreme outcomes. The arithmetic is non-intuitive but unambiguous: in a portfolio where 95% of attempts produce small returns and 5% produce 100× returns, the 5% dominate the total. The implication is structural: optimising for high frequency of small wins is the wrong objective; optimising for capture of the few large wins is the right one. This reframes how to evaluate effort: many failures aren't a sign of bad judgment; they're the cost of being in the game where the tails compound.

Conditions

Holds when:

Fails when:

Evidence

"tail events (a tiny fraction of outcomes) drive the majority of results, which means you can be wrong the vast majority of the time and still succeed if you get the few big decisions right."

— see raw/expert-content/experts/morgan-housel.md line 18.

Signals

Counter-evidence

The tail-events frame is sharpest in high-variance domains and least applicable in steady-state operational ones. Companies that try to apply power-law thinking to operational excellence (where consistency matters) damage their reliability. The discipline is matching the framework to the actual outcome distribution: empirical, not assumed.

Cross-references

Open the interactive view → View original source → Markdown source →