How does "Know and mark the edges of your circle" work?

An unmapped edge is invisible and therefore crossable without warning. Explicitly marking the boundary — "I know retail banking well; commercial real estate I only know the surface of" — creates a cognitive speed bump before decisions in adjacent territory. The edge is where confidence is highest and accuracy is most likely to have diverged, making it the most dangerous point in the circle.

How does "Stay inside for high-stakes, irreversible decisions" work?

The circle of competence is most practically valuable not as an abstract map but as a decision rule: when the stakes are high and the decision is hard to reverse, restrict independent judgment to domains where you are genuinely expert. Outside the circle, the expected value of autonomous decision-making falls — you are not better than average, and you may be worse because you cannot accurately estimate your own error rate.

How does "Expand the circle deliberately, not aspirationally" work?

The circle only expands when genuine feedback corrects your model — when you have been wrong, understood why, and revised. Reading widely, taking courses, or feeling interested in a domain does not expand the circle; making real decisions, getting accurate feedback, and updating from it does. Aspirational expansion ("I’m learning finance") conflates intention with achievement.

How does "Ask 'What am I missing?' before major decisions" work?

Confidence is a feeling generated by familiarity and fluency, not by accuracy. High confidence inside a domain can mask the unseen edges — what you would have needed to know to include the unknown unknowns. The question "what am I missing?" is a forcing function that recruits deliberate reflection to counteract the automatic confidence signal.

How does "Admit uncertainty out loud, especially in your own domain" work?

Admitting uncertainty has two effects: it signals epistemic humility to collaborators, who then contribute information they might have withheld from someone projecting certainty; and it prevents the speaker from being locked into a position by public commitment. The social cost of "I don’t know" is usually lower than the decision cost of false confidence.

Buffett & Munger

What is the circle of competence and how do you use it to make better decisions?

Know what you know, know what you do not know, and act accordingly

Short answer

Warren Buffett and Charlie Munger’s circle of competence is the idea that each person has a domain where their knowledge is genuinely deep, and that the highest-quality decisions come from staying inside that domain — or clearly knowing when you have stepped outside it. The hard part is not identifying the circle but accurately estimating its size.

Buffett has described the circle of competence as one of the most important ideas in his investment approach: 'What counts for most people in investing is not how much they know, but rather how realistically they define what they don’t know.' The principle extends well beyond investing — any domain where overconfidence is costly benefits from a clear-eyed assessment of where genuine expertise ends and where guesswork begins. Here are the practices that make this mental model operational, with honest evidence.

The practices (6)

Why it works

Most people overestimate how much of their knowledge is deep versus thin or borrowed. The Dunning-Kruger effect — the well-replicated finding that people with limited knowledge in a domain rate their competence higher than those with genuine expertise — operates most damagingly at the boundary of the circle. An honest mapping process surfaces the inflated portions before they drive bad decisions.

How to do it

1List five domains where you make regular decisions and rate your confidence in each.
2For each domain, try to teach the fundamentals to someone else — where you hesitate or simplify, the knowledge is shallower than it feels.
3Identify the last time you changed your mind significantly in each domain; absence of updates can signal overconfidence.
4Ask a respected peer in each domain to rate your knowledge — calibrate against their view.

Evidence

Observational

The Dunning-Kruger effect — overestimation of competence by those with limited knowledge — is well replicated, though its exact magnitude and mechanism remain debated. The core finding that metacognitive accuracy improves with expertise is robust.

Honest caveat: Recent reanalyses suggest the original effect size was partly a statistical artifact; but the directional finding — that low-competence people overestimate and experts are better calibrated — remains broadly supported.

— Kruger & Dunning (1999), "Unskilled and Unaware of It", Journal of Personality and Social Psychology

Common mistake: Mapping your circle based on how much time you have spent in a domain rather than how often you have been right — tenure is not the same as competence.

Go deeper on this practice →

Practice this with IX Coach

Reading about a practice changes nothing on its own. IX Coach turns these into a guided, adaptive routine — discerning where you are in real time and walking the practice with you, session after session.

Start with IX Coach →