R — Original Text (Reading)
"Circle of competence" means the range within which you can judge the discounted future cash flows. Knowing how big your "circle of competence" is matters far more than how big the "circle of competence" itself is. I think margin of safety should actually refer to the circle of competence, not just the price. The circle of competence is: be honest with yourself—know what you know, and know what you don't know. With that attitude, if you can understand something, then it is within my circle of competence; otherwise, it is not.
— Duan Yongping, Chapter 1 Investment Philosophy / Chapter 3 On the Circle of Competence and Principles
I — Methodological Skeleton (Interpretation)
The circle of competence is not a "familiar industry"—it is the approximate range within which you can judge a company's future 5-10 year cash flows. Industry experience does not equal the circle of competence: a doctor understanding pharmacology does not mean they can judge the commercial prospects of an innovative drug.
Duan Yongping made a fundamental redefinition of margin of safety: margin of safety is not a price discount (buying cheap enough), but the depth of your understanding of this company's business model. From a 10-year perspective, a purchase price 30% above the lowest point has little impact, but if you misjudge the company, you lose 100%. Therefore position size should reflect depth of understanding, not how cheap the price is.
Knowing the boundary of your circle of competence is more important than the size of the circle itself. People who don't know the boundary operate outside their circle of competence without realizing it—this is far worse than people whose circle is small but whose boundary is clear.
The only criterion for judging whether something is within your circle of competence is: can you explain in one sentence what this company will rely on to make money in the future, how much money it will make, and why competitors cannot take it away? If you can't explain it clearly, it's not in the circle.
The circle of competence is very hard to expand through reading. Truly understanding a business often takes many years. Expanding the circle of competence is not necessarily wise—within the scope you understand, there are already more than enough opportunities.
A1 — Application in the Book (Past Application)
Case 1: NetEase (within the circle of competence — 100x return)
- Problem: In 2001-2002, NetEase's stock price fell to $0.8-1, and the market generally was bearish on the gaming industry's prospects
- Use of the methodology: While making Subor game consoles, Duan Yongping accumulated more than a decade of understanding of games—he did not need to read financial reports, but instead played NetEase's games himself to verify his judgment. He knew the gaming market was far larger than the market expected; what was uncertain to others was very certain to him
- Conclusion: NetEase was within his circle of competence, his depth of understanding was sufficient, and he could take a heavy position
- Result: Bought at $0.8, held most of it until selling at $120-140, achieving a return of more than 100x. The fundamental reason he could hold on was truly understanding the business, not willpower
Case 2: FRNT Airlines + UNG Natural Gas ETF (outside the circle of competence — major loss)
- Problem: Holding a large amount of cash, seeing a "cheap" opportunity and wanting to act
- Use of the methodology (failure): FRNT — bought into the "logic" that oil would fall and airlines would benefit, but didn't understand airlines' credit risk and management behavior. UNG — knew natural gas prices were below long-term cost, but didn't understand the ETF's structural decay and time decay.
- Conclusion: Neither target was within the circle of competence. It seemed to have logic, but in reality it was "thinking you understand, when you actually don't yet know what you don't know."
- Result: FRNT exited at a loss after the CEO suddenly filed for bankruptcy; UNG was held after a loss in hopes of breaking even, and by the time it was finally cut, the loss had expanded nearly 4x. Natural gas later rose 50%, but UNG's structural flaws still left holders losing money. Duan Yongping admits: the money lost amounts to hundreds of millions, and every single loss came from violating "don't do things you don't understand."
A2 — Trigger Scenario (Future Trigger) ★
In what situations would a user need this skill?
- Substituting industry experience for business judgment — The user is a practitioner in some industry (doctor, programmer, engineer) who believes "I understand this industry, so I can invest in it," but in reality can only judge the technical/professional level and cannot judge the 5-10 year business outlook (pricing power, competitive landscape, profit model).
- Wants to act after researching for a while — The user has spent several weeks researching a target and listed several positive reasons, but when asked "how will this company make money 5 years from now, and how much," cannot answer, or can only repeat research report content rather than give their own judgment.
- Wants to buy something they don't understand because it's "cheap" — The user finds that a target has a very low PE or has dropped a lot in price and thinks "the margin of safety is large" so they should buy, but cannot clearly explain the company's business model and future cash flows.
Language signals (activate when these appear in the user's words)
- "I've worked in this industry for X years, I should be able to understand it."
- "After half a month of research, I feel I understand it pretty well, getting ready to act."
- "This one's PE is only 8 / it's down 60%, the margin of safety is large."
Distinction from adjacent skills
- Difference from
stop-doing-list: SDL is a negative list of "what not to do" (no shorting, no margin); circle of competence is the positive definition of "what range you can act within." SDL is the outer line of defense, circle of competence is the core positioning. - Difference from
good-business-identifier: The good business identification method judges "whether this business is good" (business model moat); circle of competence judges "whether I am qualified to judge whether this business is good" (cognitive boundary). Pass the circle of competence gate first, then analyze with the good business framework. - Difference from
dcf-thinking: DCF thinking is a valuation tool (roughly estimating future cash flows); circle of competence is the precondition for using DCF — only for companies within your circle of competence are you qualified to make a DCF judgment.
E — Executable Steps (Execution)
Once the skill is activated, the agent should execute the following steps:
-
Test understanding: use the "one-sentence cash flow test"
- Require the user to answer in one sentence: how will this company make money over the next 5-10 years, roughly how much will it make, and why competitors cannot take it away.
- Completion standard: the user can give a clear, coherent, unambiguous answer, and the answer does not rely on research report wording or third-party views.
- Stop condition: if the user can only repeat information and cannot make an independent judgment, directly determine that it is "outside the circle of competence" and skip to step 3.
-
Test boundary awareness: use the "counter-argument test"
- Ask three follow-up questions: (1) What might you have overlooked? (2) Under what circumstances would you admit you were wrong? (3) Can you name at least 3 specific risks that could cause this investment to fail?
- Completion criteria: The user can proactively identify their own cognitive blind spots and can give specific, non-generic risk scenarios
- Stop condition: If the user cannot name risk points or believes there is "basically no risk," this itself is a signal of insufficient understanding
-
Decide action based on level of understanding
- Deep understanding (passes both tests) → Decide position size within the scope of understanding (the higher the understanding, the larger the position can be; this is "margin of safety = level of understanding")
- Surface understanding (passes the first but not the second) → Probe with a small position or do not invest, and admit "I am not sure"
- Not within the circle of competence → Do not invest. Duan Yongping's original words: "If you feel that in the end there really is no company you can understand, then do not touch anything, and your investment performance can rank in the top 10-15% in the world"
- Completion criteria: The user gives a clear conclusion on "invest / do not invest / how much to invest" based on level of understanding, rather than based on price, emotion, or others' advice
B — Boundary ★
Do not use this skill in the following situations
- Pure information inquiry — The user only wants to understand basic information about an industry or company, and it does not involve an investment decision
- Users who already have a mature investment framework are doing marginal optimization — If the user already has a self-consistent decision system and good long-term returns, the circle of competence is only a fine-tuning issue for them rather than a cognitive blind spot
- Daily consumption decisions — The circle of competence is a framework designed for long-term investment decisions and does not apply to small consumption scenarios such as which phone to buy or which restaurant to choose
Failure modes the author warns about in the book
- "Thinking you understand when you actually do not" — Duan Yongping's own biggest failures (FRNT, UNG) both stemmed from this. He has repeatedly emphasized "we do not know what we do not know"—you think you know, but you actually do not yet. This is the most dangerous trap in the circle-of-competence framework, because the person involved is completely unaware
- Using "expanding the circle of competence" to rationalize overstepping — Duan Yongping observed that many people go outside their circle of competence to do things, then comfort themselves by saying "expanding the circle of competence requires paying tuition." He calls this mentality "being knocked back passively is far worse than staying inside actively"
- Overconfidence among highly intelligent people — "People with high IQ may often easily step outside their circle of competence." Smart people more easily convince themselves "I can understand this," and as a result fall even harder outside their circle of competence
The author's blind spots / limitations of the era
- Can the boundary of the circle of competence be known? — Duan Yongping says "knowing the boundary is more important than the size of the circle of competence," but in reality the boundary is blurry. He himself also admits "you probably only understand after losing money," which means truly identifying the boundary often requires paying a price rather than relying purely on introspection
- The assertion that the circle of competence is fixed by age 40 — He quotes Munger, "I have never seen a value investor under 40," implying that the circle of competence is difficult to establish early in life. This may be overly pessimistic for young people and ignores the possibility of transferring experience across industries
- Cases are concentrated in consumer and technology — Duan Yongping's circle of competence (gaming, consumer electronics, baijiu) happened to be the best-performing industries of the past 20 years. If his circle of competence had happened to fall on cyclical stocks or finance, the strategy of "staying within one's circle of competence" might have performed completely differently
Easily Confused Adjacent Methodologies
- "Familiar Industry" vs "Circle of Competence" — A familiar industry is perceptual and at the level of experience; the circle of competence requires being able to judge future cash flows, and is rational and at the level of analysis. Duan Yongping clearly distinguishes: "Uncertainty about others is often very certain to oneself"—what a true circle of competence brings is certainty, not merely a sense of familiarity
- "Margin of Safety = Cheap Price" vs "Margin of Safety = Depth of Understanding" — Traditional value investing defines the margin of safety as the extent to which the purchase price is below intrinsic value. Duan Yongping redefines it as depth of cognition—from a 10-year perspective, paying 30% more is irrelevant, but misjudging a company loses 100%. This is a fundamental revision of Graham's framework
Related skills (to be filled in Phase 3)
- depends-on: [ownership-mindset]
- contrasts-with: []
- composes-with: [good-business, concentrated-investing]
Audit Information
- Verification passed: V1 ✓ / V2 ✓ / V3 ✓
- Test pass rate: Pending Phase 4
- Distillation time: 2026-04-16