Cedric Chin
Experts can't explain what they do — so he stopped listening and went after the eyes

1. One strange thing

Chickens don't have penises.

That anatomical fact created the most expensive job in the egg industry. The sooner a chick's sex is known, the sooner a farm stops buying feed for males — it is the line between a backyard operation and an industrial one. Expert sexers, in the words of one report Chin quotes, are "guarded like state secrets and paid like local gentry," starting at $60,000.

Their throughput: a thousand birds an hour at better than 98% accuracy. One Hugh Grove of Oregon reached 1,500 at his peak. Half a second per chick.

In December 2018 Chin put the question in its most awkward form:

How do you get from "it's yellow and it doesn't have a penis, what?" to an average of 0.5 seconds per chick?Chicken Sexing and Perceptual Learning as a Path to Expertise, 2018-12-11

If you guessed the answer involves a lecture hall and an expert explaining how to squeeze —

well, you're wrong. Expert sexers can't explain how they know. They just … do. And the people they train can't explain it any better than they do.Ibid.

What the Japanese worked out in the 1920s is this: put the trainee in a room with ten thousand chicks, have them pick one up and put it in a tray — a pure guess — with an expert beside them saying "no" when it's wrong.

A few thousand chicks in, the trainee starts beating chance. The course at the Zen-Nippon Chick Sexing School runs two years. After two years: 98%.

And they still can't say what they're seeing. In Chin's phrasing: your conscious mind thinks it is picking at random, but an invisible force guides your hand to the right tray.

Eight years and four hundred posts are, at bottom, one long dig into that well.

2. Who he is

Cedric Chin is Malaysian, raised in Kuching, a city of 300,000 on Borneo. He went to university in Singapore and it went badly: he retook first-year maths from his first year to his last, the classmates around him getting younger each time, and finally scraped a C on a resit and graduated a year late. His own explanation — a mediocre high school in a small Malaysian city gives you a bad foundation, plus an unwillingness to admit he had a problem.

He is not an academic. He ran a Singaporean company's engineering office in Vietnam for three years selling point-of-sale systems, then built a business of his own to $4.5m in revenue between 2014 and 2017. That identity runs through everything; he nails it down in the most important post in the tacit knowledge series:

This blog is a practitioner's blog: it is interested in what is useful, not what is 'true'.Copying Better: How To Acquire The Tacit Knowledge of Experts, 2020-06-16

The blog opened in May 2018 as Commonplace and became Commoncog around 2021, when the self-description shifted from career advice for knowledge workers to "a publication dedicated to the serious study of business," and membership and the Case Library went live. It is still running — the most recent post at review time is 16 September 2026. There is a baby in the picture: the best piece of self-correction he wrote in 2026 opens with him travelling with an infant, sleep-deprived, and falling ill on his return.

3. What he's been wrestling with

His root question didn't come from a book. It came from a person.

In his first year of secondary school he met Samuel, whose father was a traditional Chinese businessman importing pharmaceuticals into Kuching for a handful of local pharmacists. Over the years the business grew outward: small property holdings, food and beverage outlets, shops selling phones and data plans. Anything underperforming got shut down fast. Every year Chin came home and found the empire had expanded — in his own words, "inch by hard-won inch." Samuel was being groomed to take it over.

In Malaysia and Singapore, "Chinese business acumen" is a story everyone grows up inside. Chin is clear-eyed about what it does: it is a group-level superiority complex, and its function is to provide an exit. A Chinese kid who loses in the education system still has a route to immediate respect — going into business. Succeed and you're celebrated; fail and it was still a respectable thing to attempt.

So the question became: what exactly is Chinese businessman wisdom, and how do you get more of it?

He had standing to ask. Three years in Vietnam put him up close to these operators — some were clients, some were competitors racing him for customers ("We also drove a few of them to bankruptcy, but that's another story for another day"). His first hypothesis was rationality. The best of them really were relentlessly pragmatic; he remembers a client past fifty who extracted a pile of free add-ons so skilfully that nobody noticed for months.

But rationality didn't cover it. He split the puzzle into three observations, and each one kills an obvious answer:

First, the most successful Chinese businessmen were often the least educated. If intelligence decided it, the educated ones should have done better. They didn't.

Second, the group was more superstitious than most, and it seemed to make no difference to how well they ran their businesses. If rationality decided it, the superstitious should have lost. They didn't.

Third, the great majority didn't think far ahead at all. They optimised locally, watched the bottom line, and killed whatever wasn't performing. If strategy didn't matter, these local optimisers should have run away with it. They didn't either.

All three at once means intelligence, rationality and strategic thinking — the three nearest answers — are all wrong. Chin wrote himself an unusually honest note:

If I succeed in business, later in life, you may point to these posts as a record of what I got right. But if I failed, then these posts would be a record of all the wrong things I believe that ultimately did me in at the end.The Chinese Businessman Paradox, 2018-09-14

There is a more private floor under this question, which he later dug up himself. In the 1997 Asian financial crisis he was nine and his father was laid off when the company shut its operations in his hometown. His parents were under enormous strain and fought constantly; his grades collapsed; he got yelled at over nothing. A few years later the family sold the house he grew up in. His father tried business consulting, ran a car workshop, joined an MLM with his mother — none of it worked — and only years later found a stable job he held until retirement.

Looking back, my obsession with career moats probably started with the 1997 Asian financial crisis.A Personal History of the Career Moat, 2019-08-19

So the idea he became known for early — the career moat — is not a motivational framework. It's what grew out of a nine-year-old watching his father lose a job. He compressed it into one hard line: job security is tied to your ability to get your next job, not keep your current one. And he explains why that still sounds novel: social beliefs change more slowly than economic reality. His parents still believe security means a safe, stable job at a large, well-respected company.

4. His four ideas, and how he walked into each one

(i) "Experts can't explain themselves" isn't modesty — it's a usable signal

What made him notice the problem was a colleague named Hieu.

Hieu was his technical lead, and he had an uncanny knack: sit through a requirements meeting and, within minutes, sketch the simplest possible program structure with the fewest moving parts — while leaving room for the changes any software project inevitably brings. That sketch was usually what got built. Chin's own designs always needed rework.

He spent years trying to get it out of him. The conversation always went like this:

"Well" Hieu would begin, "When you hear there is an external API, you should focus your program around that because there is a lot of risk there." — "Yeah but then why didn't you worry about the calendar API?" — "Oh, because I've worked with it before and I think it is easy to implement." … "But how come you didn't worry about the inventory API? We've never integrated with that before." — "Ya that one not that important now I think. The client might change it later."Why Tacit Knowledge is More Important Than Deliberate Practice, 2020-06-09

Push further and you hit bottom: "Ahh, it just feels right. Do it long enough and it'll feel right to you too."

Chin's move was to refuse to treat this as Hieu being inarticulate. He treated the phenomenon itself as data, and found it generalises into a form: experts in any sufficiently complicated domain explain themselves as "do X — except when you see Y, then do Z, because A. And if you see B, then do P. But if you see A and C but not B, then do Q, because reason D."

Hence the diagnostic: a long list of caveats means you are looking at tacit knowledge. His prettiest corroboration is Atul Gawande on appendectomy — just for freeing an inflamed appendix from everything it has stuck itself to, Gawande lists a cotton-tipped pusher, electrocautery, a hook, scissors, sharp-tip and blunt-tip and right-angle dissectors, a suction device, tilting the table so gravity helps, "or you can just grab whatever part of the appendix is visible and pull really hard." That isn't verbosity. That's a man trying to say something that can't be said.

Then he takes on the objection that usually turns into a debating game: couldn't it all be made explicit? Here he reaches for history rather than philosophy.

In the 1970s, after the Yom Kippur war, the US military commissioned studies into replacing humans with expert systems. One project was to enter Israeli aircraft maintenance logbooks into a US database with a view to replacing the maintenance officers. The young researcher sent to interview them was Gary Klein. He could fill in 80% of the forms and no more — because those men could look at a piece of damage and reason: this plane was back in service in two days; damage like this, if it had severed those lines, would take a week or two; so it can't have severed them; so here is how the damage must have propagated.

So when I wrote my report, the report was: here's what I could do that was formulaic, but here are the critical incidents that I collected, so you can understand their expertise. And as a result the Air Force decided they could not build an expert system to replace these guys, and they had to keep paying them.Gary Klein, quoted in ibid.

The story matters because it converts an endless online argument about whether tacit knowledge exists into a bill that has already been settled: somebody spent real money trying, failed, and kept paying the humans. Chin's handling of the argument is blunt to the point of rudeness — if you want to insist that all tacit knowledge is in principle explicable and therefore doesn't exist, that is pedantry he isn't interested in:

We should act as if tacit knowledge were a fact, because it is more useful to think about ways to gain that tacit knowledge directly, instead of hoping for some breakthrough to make tacit knowledge explicit.Ibid.

This swap is the hinge of the whole blog: replace an ontological question (does it exist?) with an operational one (what do I do about it?). Once you do, the problem becomes tractable. If large parts of any skill tree are tacit, then they can be hunted — that person has it, that person is really good at it, how can I have it too?

(ii) Deliberate practice fooled everyone, and he found out by failing at it

If tacit knowledge is learned through imitation, apprenticeship and osmosis, the obvious next question is whether you can get better at that.

Most people answer "deliberate practice." Chin did too, and he genuinely tried. He had read Cal Newport in So Good They Can't Ignore You: deliberate practice is the best known method for building expertise — a sentence Chin later goes out of his way to call perfectly accurate. He practised.

Four years. It kept not working.

Then comes the single most important move in his method:

Eventually, I got fed up enough to dig into the actual literature — not the blog posts or the popular science books that cited K. Anders Ericsson's research, but the actual books and papers Ericsson had written himself. And what I found surprised me.Practice as the Bar for Truth, 2020-08-03

What he found: "deliberate practice" is simply a name for the forms of practice Ericsson found in easily measurable domains that already had an established pedagogy — and it has to be done under an instructor.

Why those domains? Chin's explanation is unsparing and I think correct: Ericsson was an academic, and academic incentives reward novel results that survive peer review. Music and chess were in because their training methods are well established and success is unambiguous — tractable. Business and jazz were out because neither had established training methods — not tractable.

That is the caveat the popularisation sanded off. Over thirty years, "in domains with an established pedagogy, deliberate practice is the most effective form of practice" became "deliberate practice is the most effective form of practice." And so programmers, designers, founders and investors the world over have been jamming a key cut for pianos into their own locks.

Chin adds one more cut, and it's a hard one. Having laid out the narrow definition in Peak, Ericsson then suggests the ideas may apply to less pedagogically established fields — which Chin calls a cop-out. And Ericsson cannot have been unaware of NDM: he was one of the editors of the Cambridge Handbook of Expertise and Expert Performance, alongside many names from that community.

So the road forks. In a domain with an established pedagogy, use deliberate practice. Without one — and most modern professions are without one — go to NDM. And the learning path NDM describes is almost embarrassingly old-fashioned: find a master, work under them for years, and learn by emulation, feedback and osmosis. (His example is Buffett's years under Benjamin Graham.) What NDM contributes is making that ancient process more efficient.

In 2020 he made a prediction worth keeping an eye on. Ericsson took about thirty years to reach mainstream consciousness, largely via Outliers. So what about NDM? "Perhaps we'll talk about tacit knowledge in 2030 the same way people talk about the '10,000 hour rule' today."

(iii) Business is a closed system, and the good ones carry the same triad

By now he had a tool (NDM) but not the content: what, specifically, is there to know about business? That is exactly what the 2018 paradox left unanswered.

In 2021 he ran into Lia DiBello, an NDM researcher who specialises in extracting the tacit mental models of business expertise. His own reaction to the first passage he read was that the claims "sound ridiculous":

After several years using different methods, we noticed that highly talented business performers are very similar to each other. That is, they have decoded the domain of business much like chess masters had decoded chess. … as a domain—business itself is an orderly closed system of relations between principles, and that so-called intuitive experts in business have an implicit grasp of this.Lia DiBello, The Oxford Handbook of Expertise, quoted in Every Great Business Person Has the Same Mental Model of Business, 2021-07-20

In a four-year study funded by the National Science Foundation, DiBello studied repeat successes — leaders who kept winning in hard markets, some of them chairmen of large corporations — and reported finding a shared mental model among all of them.

The model is a triad: (1) factors involved in effective operations; (2) forces influencing the market; (3) those driving business finance and economic climates. Across her papers she names it differently — supply/demand/capital, or leadership/strategy/finance.

The vertices are unremarkable — everyone knows a business has operations, a market and a balance sheet. The entire weight of the model rests on the dashed lines: when one leg moves, what happens to the other two. Capital conditions change — what does that do to competition? Competition shifts — what must operations do in response? Experts pick up environmental cues long before novices, then forward-simulate the other two legs.

From which comes the sharpest implication in the whole body of work:

In contrast, competent managers (one level down from experts) tend to be very talented in only one or two of these areas; however, they often do not understand the dynamics between these areas as well as the "superstars" do. Competent managers are likely to be very successful when larger market or economic trends are favourable to their specific skills.Ibid.

In plain words: in a tailwind, mediocre and excellent look identical. And you can't separate them by results, because results are the thing the wind is pushing.

The best evidence for this comes not from a researcher but from a man it frightened. DiBello describes a medical device company whose chairman and acting CEO asked her to evaluate his senior staff for succession. Over the four months she worked with them, the company grew from $300 million to $400 million, with a pending acquisition meant to take it to $600 million. Everything was going beautifully. And he wanted her there because:

He was afraid that a favourable market might be masking a lack of skill—and cognitive agility in particular—among his senior staff.Ibid.

Cognitive agility is the second component beside the triad: the extent to which someone revises their read of a situation when the data says conditions have changed. Its converse is cognitive rigidity — being dominated by a rigid frame that filters out new and possibly relevant information, creating blind spots.

Chin's reaction to this passage, in his words, was that his eyes basically popped out of his head. He had spent years writing the same thing from an operator's seat: when a market changes, no framework captures everything you're seeing as you brush against reality. He had even coined a trait of his own, "Dismissively Stubborn," for the sort of person he had learned to avoid in startups.

Then he discovered DiBello had got there a full decade ahead of him, and more thoroughly.

The triad's virtue is that it immediately pays off old debts. Why can't fresh MBAs grasp what cash flow means inside a specific business? Because they haven't internalised the relationships between market conditions, operational realities and the financial concepts they learned in school. (DiBello's data backs this: business students who had never run anything scored much lower on the same instrument.) Why does Buffett say he's a better investor because he's a businessman and a better businessman because he's an investor? Because investing and operating shore up different legs. Can consultants or ex-VCs become good operators? Now you can ask it precisely: consultants are better equipped on demand with some finance; VCs on finance with some demand; how well they do depends on how fast they level up on supply and internalise the couplings.

(iv) People who don't run businesses think business is profits; people who do know it's cash flow

He dug out the most counter-intuitive leg separately, and the way he did it is characteristic: it starts with an argument he couldn't win.

In autumn 2020 a friend sent him an essay called "Startups Shouldn't Raise Money." Chin thought it was well built — every step followed cleanly from the last, argued from first principles, so no individual clause was refutable.

He also thought it was wrong. His friend asked how. He opened his mouth and stopped. He agreed with every axiom and nodded at every intermediate proposition.

He chewed on it for months. The description he eventually landed on for this class of error: "perfectly rational, logically constructed, and not really wrong — but not as useful or as powerful as some other framing." The diagnosis: the author started from a deficient set of axioms. The essay treated startups as a game about profit and incentives.

people with limited understanding of business think that business is all about making profits. But those who actually run businesses know that running a business is all about managing cash flows.The Games People Play With Cash Flow, 2020-12-09

This is the thing he first heard from the Chinese businessmen, who like saying "pay other people as late as possible" and "cash flow is king." Chin points out their terminology is wrong in a way that matters — search the business literature with those keywords and you'll find nothing useful, because what they mean is not cash flow but free cash flow. And the intuitive definition he gives beats any textbook:

FCF is the amount of money that you can take out of a business without affecting its future prospects.Cash Flow Is King, 2019-10-01

He turns it into a picture: if a business makes $1 million in profit and $800k of that is free cash flow, you can take the $800k out and set it on fire, and the business will still produce at least $1 million next year.

And the rest? In a company like Intel, far more than you'd think — don't put it back into fabs and process, and next year's million evaporates. So high profits do not mean you have money. A person who only counts profit will never see this leg at all.

Then he lays out John Malone's story as the extreme case of what someone who does see it can do. On April Fool's Day 1973, Malone took over a near-bankrupt cable company, TCI, carrying debt at about 17 times revenues. For his first couple of years he flew to New York every few weeks, hat in hand, renegotiating covenants and begging extensions; at one meeting he threw his keys on the table and threatened to walk, and the bankers capitulated.

But Malone had read the industry's structure before he took the job: highly predictable subscription revenue; cable franchises that amount to a legal right to a local monopoly; and tax characteristics so favourable that you could shelter cash flow by building with debt and depreciating aggressively, then sell the system to another operator and let the depreciation clock start over. That is all three legs moving at once — market structure, operating expansion, capital instrument. Back at Bell Labs he had already advised AT&T to increase its debt load and shrink its equity base through buybacks. The advice reached the board and was ignored.

It is also his answer to where moats come from: temporary competitive advantage can emerge from any of the three legs — operational excellence (supply), a market opening or repositioning (demand), or exploiting a change in the capital environment nobody else in the industry has noticed (capital), which is exactly what Malone did with debt and tax shelters.

5. How he thinks

His method travels better than his conclusions. Eight years of reading yields five rules, each with evidence you can point at.

One: one ruler, and it's the crudest one available. He says all the rigour on the blog compresses to "use practice as the bar for truth." Most people use analysis as the bar — checking an argument for internal consistency, constructing calibrated beliefs. His reason is unromantic: "Analysis is a lot of work. But practice is a shortcut. You apply something to your life and if it works, it's true." The ruler's hardest cut lands on growth mindset: he spends pages arguing the research may not hold (shrinking effect sizes, results too neat to trust), then admits it broke the four-year deadlock with his maths classes — he had made "I'm bad at maths" part of his identity, and dropping that let him go ask the freshmen for help and scrape a C. The conclusion: "something doesn't have to be true scientifically for it to be useful for you."

Two: go back to the primary source on anything that reaches you through a citation. Four years of failed practice broke only when he stopped reading books that cited Ericsson and read Ericsson. This sounds like standard advice, but note where it sits in his story: it isn't a virtue, it's a stop-loss — something exasperation forced on him.

Three: hostility toward anything that sounds insightful, including his own writing. In 2019 he wrote a whole essay on this, opening with Ida Tarbell's canonical takedown of Rockefeller — and Chernow's finding that Tarbell's brother was treasurer of Standard Oil's biggest competitor, fed her material, and even vetted her manuscripts. His thesis:

The best online writers are able to make something sound insightful — regardless of whether it's true, or whether it's useful. … Some ideas sound insightful because they are true. But not all true ideas sound insightful. A true idea that is commonly accepted can sound trite and obvious: we call those clichés. The job of a good writer, then, is to present some truth in a way that doesn't trip our cliché triggers.Beware What Sounds Insightful, 2019-09-24

Then he does something I've rarely seen a writer do: he goes back and dismantles the tricks in the essay you are currently reading. He points at his own line — "if Instagram has changed the language of photography, then we should also expect the emergence of Twitter, Medium and Google to change the tenor of writing" — and says it is an obvious truth rewritten to sound insightful; writers have always optimised for attention, and you knew that; the Instagram reference is what threw you. He also notes he wrote "optimised for attention" instead of "liked to be read" purely because it sounds more sophisticated.

Four: prefer repeat successes to single successes. This one hides in his taste in subjects. DiBello studied chairmen who won repeatedly; his own twenty-year case study is Samuel's father expanding inch by inch; he wrote a dedicated post telling you not to take generic business advice from VCs. The logic is consistent — one success can't be separated from luck, repeated success carries signal.

Five: when wrong, dismantle it in public, and don't quietly patch. This is his best piece of 2026. He had commissioned a case on the quartz crisis — how Swatch saved the Swiss watch industry, the standard account for decades and the one the Swatch Group itself pushed. The case was written, published, and he was about to record a reaction. Then he travelled with a baby, got sick, and while convalescing picked up Pierre-Yves Donzé's A Business History of the Swatch Group.

And then I realised this entire narrative was wrong.Did Quartz Really Kill the Swiss Watch Industry in the 70s?, 2026-06-25

Wrong in two large ways. First, the Japanese were beating the Swiss with mechanical watches before they got quartz working — so the real illness was production technology, which quartz merely accelerated, and framing it as product disruption gets the whole case backwards. Second, the cash that funded Hayek's acquisition spree came mostly not from Swatch but from the luxury-positioning play Jean-Claude Biver proved out at Blancpain. Reframed, the case isn't a story of technological disruption at all; it's the story of global luxury emerging as a business.

His handling is worth copying: he put the wrong version on Google Docs, wrote "which you may read, if you so desire," and rewrote it. And he reminded readers why they were reading the case at all — "We are not reading this for the sake of history (although that's good to do!) We are reading this to figure out how Hayek and Biver and the rest made sense of the facts they saw during the crisis. That way we can use this fragment in our own sensemaking."

Which connects straight to where he has arrived in 2026. He now believes the core of expertise is not pattern matching but frame construction — you build a frame in your head, and the frame determines what you notice and what you ignore.

Experts can see things that novices can't. If you can teach novices to see the domain like the experts, you will accelerate expertise. … data does not exist, 'on its own', in the world.Sensemaking as the Heart of Expertise, 2026-06-12

So he collects cases not to extract lessons but to accumulate fragments — enough of them, and a person can assemble a new frame on the fly in a situation nobody has seen before. Eight years on, the room full of chicks has grown into this.

6. Where he might be wrong

First and heaviest: the evidence base for the triad is far thinner than the weight it carries. The whole business-expertise edifice rests on one researcher, and Chin says it himself: DiBello has published "a relatively small number of papers" because she spent twenty years consulting for big companies. Her core instrument, the FutureView Profiler, is a commercial product. Her subjects were leaders who succeeded repeatedly and rose very high — textbook survivorship bias. Nobody counted how many failed operators also carried a complete triad in their heads.

This bites because Chin wrote a post called "The Base Rate is a Hell of a Thing." And here is how he states his grounds for belief: "Do I buy this model of business expertise? Yes, I do. Nearly everything I know about business lines up with the model." "Nearly everything lines up" is precisely the sensation he taught us to distrust in Beware What Sounds Insightful. He credits the model with explanatory power out of the box, and a framework that explains everything usually pays for it by being falsified by nothing. His argument for DiBello's credibility is likewise indirect: she sold her soul to big business and made rich people richer, which "to my mind, makes her more, not less, believable." That is a reasonable instinct. It is an instinct, not evidence.

Second: "practice as the bar for truth" issues a permit to n=1. What makes the rule beautiful — skip the endless analysis, just use the thing — is also what makes it fragile: it has no mechanism for separating "this actually works" from "I happened to survive." The growth mindset case is his own: the science was eroding, he kept the intervention because it worked for him. He does say out loud that it might have been a placebo and that a therapist might have done the same job — the honesty is real. But hand this rule to a thousand people and you get a thousand mutually contradictory methodologies, each one "tested in practice."

Third: the 2018 Cedric Chin would tell you not to subscribe to the 2026 one. This is the truest crack in him, and he cut it himself. The essay that made his name opens without hedging:

don't read blogs written by non-practitioners, spouting insights that aren't related to their field of practice. Don't read Farnam Street. … Hell, don't read this blog — especially if your career goals diverge from mine. I have little to offer you that practice cannot.The Mental Model Fallacy, 2018-11-12

Eight years later Commoncog is a subscription business: a case library packaging the stories of Amazon, TSMC, Instagram and Ford for readers who overwhelmingly do not run those businesses. He raised membership prices in September 2026.

He has a defence, and it isn't weak: the case method doesn't codify tacit knowledge into a list — the thing he declared impossible — it delivers perceptual exposure, showing you enough situations that you grow the eyes yourself, exactly like the trainee in the chick room. He is doing the inverse of Farnam Street, not an upgrade of it. In theory the distinction holds.

Two things still hang there. One: the knife he used to execute Farnam Street — "how is this useful? How may I test this idea in practice?" — sits just as well against his own case library, except that this time nobody is holding it. Two: the incentive structure he diagnosed so precisely in 2019 — writers optimising for attention, dressing truths up so they don't trip your cliché trigger — now has a harder version aimed at him: subscription revenue. Sound insightful and people renew. In 2019 he wrote that he had used those tricks himself. He had little to lose then. He does now.

Fourth: the paywall puts the most mature work beyond outside checking. His centre of gravity since 2021 — the Case Library, the member newsletters, the accelerated expertise courses — sits largely behind it. This page is built from what is publicly readable, and the public material skews heavily to 2018–2021. Which is to say: the part of his work that most needs testing is the part outsiders can least see. That is not an accusation; it is a limit a reader should know about.

Last, and he wrote it better than I can. The 2018 Chinese businessman essay ends by saying these posts are a working notebook, not conclusions: if I succeed later in life, point at them as a record of what I got right; and if I failed, they would be a record of all the wrong things I believe that ultimately did me in at the end. This blog's biggest risk has been posted, in plain sight, on its first page the whole time.

7. Sources

• The Mental Model Fallacy (2018-11-12)The essay that made his name and the foundation of everything after: the most valuable mental models do not survive codification. Contains the line that will follow him forever — "don't read this blog."
• The Chinese Businessman Paradox (2018-09-14)The starting point. Kuching, Samuel's father, three paradoxes that kill three obvious answers, and the note he wrote himself about being judged later.
• Why Tacit Knowledge is More Important Than Deliberate Practice (2020-06-09)Part one of the tacit knowledge series. Hieu, Gawande's appendectomy, Gary Klein and the Israeli maintenance officers no expert system could replace. If you read one, read this.
• Every Great Business Person Has the Same Mental Model of Business (2021-07-20)DiBello's supply/demand/capital triad, and the chairman who feared a favourable market was masking his team's incompetence. The boldest post on the site and the one that most deserves scepticism.
• Practice as the Bar for Truth (2020-08-03)His methodology stated outright. The four failed years of deliberate practice, going back to Ericsson's own books, and why he keeps a growth mindset whose science may not hold.
• The Games People Play With Cash Flow (2020-12-09)How first principles thinking walks you off a cliff even when every axiom is true; plus John Malone building a cable empire out of debt and tax shelters.

Other quotations on this page come from: Chicken Sexing and Perceptual Learning as a Path to Expertise (2018-12-11), Career Moats 101 (2018-11-23), An Update on Perceptual Exposure as Learning Technique (2019-03-25), A Personal History of the Career Moat (2019-08-19), Beware What Sounds Insightful (2019-09-24), Cash Flow Is King (2019-10-01), Expertise is Just Pattern Matching (2020-01-21), Copying Better: How To Acquire The Tacit Knowledge of Experts (2020-06-16), How First Principles Thinking Fails (2020-12-01), Sensemaking as the Heart of Expertise (2026-06-12), Did Quartz Really Kill the Swiss Watch Industry in the 70s? (2026-06-25), Why Tactical Decision Games Work (2026-08-25), and his own curated Start Here page. Blog homepage: commoncog.com.