中文

You Write the Sign: The Same Numbers, Read Upward

JIAHON Group · JIAHON Think Tank perspective · Road to NT$100 Million, Part Two. Published by JIAHON Think Tank | Written by Marcus (Senior Vice President) | Published 20 September 2026 | Updated 2 October 2026 | jiahon.com.tw

Part One measured three roads on the same ruler, and concluded that none of them is positive. This article does not change a single number. It does one thing: it brings forward the three switches that Part One calculated but placed after its conclusion — the ticket, the margin ladder and the growth rate. All three are turned by understanding, and this era adds an amplifier that did not exist before.

In Brief

Part one measured three roads to NT$100 million—lottery, salaried saving, and starting a company—on one ruler and found none with a positive expected value for the average entrant. This sequel changes nothing in the numbers and everything in the reading. Expected value is a conditional probability, and part one had already computed the conditions that flip its sign: an industry ticket (ten years of knowing how decisions are made) turns −NT$16.19M into +NT$5.6M; the margin ladder (19% → 65% gross margin) thickens the tail 3.7× without moving the mean; and a 12% annual growth rate of the surviving company flips the sign again—the one parameter written by the operator, not the market.

We map the three switches onto the pillar order from our previous note (truth → excellence → courage → justice), then add what did not exist when the ticket was first theorised: an amplifier. Knowledge output and AI collaboration let a non-transferable ticket be written down, copied, and partly transferred—raising the zero recovery rate that made “the more valuable you are, the less your company is worth” true. A single try is still a dice roll. But the probability of each roll, and the number of rolls, are now both in the operator’s hands. All three switches stand on one precondition: a floor. p × n compounds only if no single roll can wipe the operator out—which is why growth bought with personal guarantees can hand back the survival the ticket paid for.

1. The bad news first

What Part One measured is set out here unchanged: the lottery, a fixed −45%; working, saving and investing, 44% of NT$100 million after forty years of perfect execution, 28% once marriage, a home, children and parents are added; starting a company for twenty years, an expected value NT$15.65 million to NT$16.02 million below staying employed, 24% dead in the first year, 44% left after ten, closing on average in year 3.5.

This article does not change any of those numbers. The survival rates and revenue brackets were measured, not estimated — the Ministry of Finance’s suspension files, the DGBAS census, the white paper’s revenue brackets; the amounts inside the net difference contain assumptions, clearly marked in Part One’s appendix. Change them, and nothing that follows has the right to be said.

But one thing needs to be clear first: expected value is an average, and the average person does not exist.

Part One calculated the “average entrant” — everyone who entered, prepared or not, whatever business they chose, whether or not they grew afterwards, all added together and divided. That number is useful for policy and useless for the person making a decision, because nobody is the average. Your expected value is a conditional probability: whatever the conditions are, that is what the sign is.

Averages have a second trap. Even with identical conditions, whether one person collects that average over a lifetime depends on whether there is a roll along the way that takes them straight out of the game. The physicist Ole Peters argued in Nature Physics in 2019 that economics has long treated “the average of a group” and “the average of one person’s lifetime” as the same thing. The average person does not exist; neither does the average lifetime. The end of Section 2 and Section 5 come back to this.

Part One had already calculated the conditions; it just placed them after the conclusion. Most readers stop at “none of them is positive” and never turn to the three tables behind it. This article brings them to the front.

2. Three switches that can turn the expected value of starting a company positive

None of the three switches is given by the market — the ticket is the seat you sat in, the ladder is the business you chose, the growth rate is whether you put money back in or took it out

Switch one: the ticket — it flips the sign

A: starts a company at thirty and runs it for twenty years; net difference −NT$16.19 million. B: builds a ticket in the industry on a salary from thirty to forty, starts a company at forty and runs it for ten years; net difference +NT$5.60 million. The ticket is worth NT$21.80 million, and what it buys is not a little more money but the sign. Part One put it precisely: the ticket swaps a random loss for a certain delay.

What the ticket is: knowing how other people decide — why the buyer picked this supplier, why the engineer is holding this specification, why the boss cut this budget. It has three properties: scarce, portable, redeemable. A process engineer has a ticket after ten years; a production-line operator has only seniority. It is not a question of ability; it is a question of position.

Switch two: the margin ladder — it changes the tail, not the expected value

Both listed under wholesale: anyone can enter the 19% gross-margin step, so nobody makes money there, and the ten-year multiple for reaching NT$100 million is 21×; 57% is the level a niche technical channel can reach, 67×; 65% requires customers to write you into their specifications, 77×. A 3.7× difference, and its source is neither effort nor luck; it is which product line you chose.

But this cell needs an honest statement: a 3.7× difference in the multiple, and only NT$450,000 difference in expected value. Gross margin decides how likely the “in case it works” outcome is, not whether the average outcome improves. It changes the thickness of the tail. The reason to include it is not that it flips the sign; it is that it decides how far you can go once the sign has flipped.

Switch three: the growth rate — it flips the sign, and it is the only one decided by execution

At the end of its tenth section, Part One said one cell in the model was still broken: it assumes the owner of a surviving company takes NT$1.75 million every year for twenty years, and the company never grows. Put growth in: at 8% a year, the net difference shrinks from −NT$15.69 million to −NT$11.02 million; starting at NT$1.2 million with 8%, −NT$7.05 million; at 12% a year, +NT$0.39 million — it flips positive.

Part One called this “where the whole model is fragile.” This article reads it as the hopeful cell — the lottery’s −45% is set by law, the salaried ceiling comes from the salary curve, and of the three roads only this cell holds a number filled in by the person doing the work.

Figure 1: Each of the three switches turned once — two flip the sign, one changes the tail. All figures from Part One

Table 1: Three switches — what they move, Part One’s figures, who can turn them

SwitchWhat it movesPart One’s figuresWho can turn it
TicketThe signA −NT$16.19 million → B +NT$5.60 million; ticket worth NT$21.80 millionSomeone who spent ten years in the right seat — touching “decisions,” not “execution”
Margin ladderThe tail21× → 77× (3.7×); expected value differs by only NT$450,000Someone who chose the right product line and was eventually written into the specifications
Growth rateThe signStill negative at 8% a year; positive at 12% (+NT$0.39 million)The person doing the work — putting money back in rather than taking it out

The three switches have one thing in common: none of them is given by the market. The ticket is built in the seat you sat in, the ladder is the kind of business you chose, the growth rate is whether you put money back in or took it out. Part One said starting a company is the only one of the three roads whose expected value is not fixed — these three switches are what “not fixed” consists of.

Before the switches: the floor

All three switches change the chance of winning, or how much you win when you do; none of them governs how much you lose when you lose. Part One estimated that cell: the recovery rate is only 8% for restaurants and 20% for wholesale channels; and losses in a company have no cap, because personal guarantees reach through to the individual.

The floor is not a fourth switch; it is the precondition for all three. Speaking at the University of Florida in October 1998 about Long-Term Capital Management, which had just come close to collapse, Warren Buffett put it plainly: to make money you do not have and do not need, risking what you do have and do need is not worth it, however good the odds.

Switch three needs particular care. If growth is bought with borrowing and personal guarantees, then while it flips the sign positive it eats back the survival rate that switch one bought. Neil Churchill and John Mullins wrote about this in the Harvard Business Review in 2001: a profitable company growing too fast can burn through its cash however successful its product; they called the growth rate a company can sustain from its own cash flow the “self-financeable growth rate.” What Taiwan calls “black-ink bankruptcy” is what happens beyond that line. So money has not two destinations — back in, or out — but a third: kept, as a floor.

3. The three switches against the pillar order

The previous article, “Four Pillars, Build Your Own Two First,” set out the order: truth → excellence → courage → justice. Lay the three switches over it and they line up, which also explains why the order cannot be reversed.

Truth — see clearly which step you are on.
Part One’s margin ladder is itself a table of “truth” — the margin on what you sell now can be looked up; how far your ticket has built up can be laid out and seen. Most people do not dare look at this cell, because looking tells them they cannot enter yet.

Excellence — build the ticket, climb the steps.
Going from “replying with a quote” to “replying whether it will work” is going from 19% to 40%. This stretch needs nobody’s approval and can start today; of the three switches, it is the one that takes time and is also certain.

Courage — enter only with a ticket.
Part One calculated it clearly: someone with a ticket who enters gets a positive sign; someone without a ticket who enters gets a negative sign, with a 20% recovery rate on failure. Courage is not daring to gamble; it is having the ticket in hand and daring to use it.

Justice — being written into the specification.
What the 65% step means is that the customer vouches for you through the specification. Your product has value for their production line, and they repay you through their system. It is the ultimate form of rewards matching people, and it is the step Part One said “takes ten years.”

Figure 2: The margin ladder is a projection of the pillar order — each step up, the customer asks a different question

Part One also gave the reason the order cannot be reversed: of those who enter without a ticket, 24% die in the first year. Skipping truth and excellence and going straight to courage is route A.

4. The amplifier this era adds: writing tacit knowledge down

The ticket used to exist only in one person’s head; now it can be written down and printed many times

Part One had an uncomfortable line: company valuation = transferable assets (high recovery rate) + the principal’s ticket (zero recovery rate). The ticket cannot be divided or transferred, and can only be grown yourself over ten years; a buyer can buy the stock and the customer list, but not “customers are willing to take your call.” The conclusion: the more valuable you are, the less your company is worth.

That line had no solution in the era when the ticket was first theorised. Now it does, and only in the last few years.

The essence of the ticket is “knowing how other people decide.” That kind of knowledge used to exist only in one person’s head, and every time it was redeemed the person had to be there. The philosopher Michael Polanyi called it tacit knowledge: we know more than we can tell. Ikujiro Nonaka and Hirotaka Takeuchi, studying how Japanese companies innovate, called the act of writing tacit knowledge into words so that an organisation can share it “externalisation.” Now it can be written down: an article on “when to replace a squeegee” is a small piece of ticket moved from a head into a knowledge base; accumulate a few hundred of them, and for the first time the ticket becomes an asset that is copyable and partly transferable — customers read it themselves, the engineering team answers from it, a second location opens by following it. The recovery rate moves up from zero.

The second layer of the amplifier is AI collaboration. Writing a ticket down used to take a department; now one person with AI can write that volume. This is a window: within a few years “able to work with AI” will be no more of a differentiator than “able to use Excel” — but what is written during the window is still there after it closes.

Figure 3: The amplifier — the ticket moves from one person’s head into a company’s knowledge base

One thing to state honestly: knowledge gets copied, and what copies it is AI. Feed in a few hundred articles and what comes out is the average answer anyone can get. What cannot be taken is the shop floor tied behind each article: the samples sent, the trials run, the part that was written into a specification. So the amplifier does not amplify knowledge itself; it amplifies the combination of “ticket plus shop floor.” Knowledge without a shop floor is content handed to competitors for free.

5. A single roll is a dice roll; a lifetime is a probability: every roll must be one you can afford to lose

Part One’s 44% is the survival rate of one roll of the dice, and −NT$16.19 million is the net difference of one attempt at a company against staying employed. But people do not roll only once.

The ticket and the ladder change p — the chance of success each time. Knowledge output changes n — the number of rolls: each article is a zero-cost contact, and when a customer calls after reading, the chance of closing that call is no longer 30%. Opening locations is addition: each location needs people, stock and local trust, and each one added is another set of dice.

Raise 30% to 80% and a single attempt can still come up empty; over a hundred attempts, thirty deals versus eighty is a gap of nearly three times. Part One’s expected value is for a single attempt; the expected value of a lifetime is p times n, and neither number is given by the market.

But p times n has a precondition: every roll must be one you can afford to lose. Taleb gives an example in Skin in the Game: a hundred people each go to a casino once, and you can calculate their average result; one person goes a hundred times in a row, and if they lose everything once along the way, the average after that no longer concerns them. The numbers make it clearer: with a one-in-six chance of being knocked out on each roll, the chance of still being in the game after 6 rolls is 33%, and after 20 rolls under 3%. So n amplifies only dice you can afford to lose: an article loses only working hours, so you can keep rolling it; a new location must first be shown not to drag down the core business if it fails — losses capped, no cross-guarantees — and only then is it a die; otherwise it is a bullet.

p also has to be shown not to be luck. In 1984 at Columbia University, rebutting the claim that “investment performance is just coin-flipping,” Warren Buffett used an argument: if the winners of a national coin-flipping contest came disproportionately from one small village, that would not be luck; the village would be teaching something. The p of the ticket and the ladder is the same: one company saying it can reach 57% is a story; when the same kind of channel can all reach it, that is the village.

Table 2: What Part One measured and what this article says — the same numbers, two time scales

Part One (single attempt)This article (lifetime)
Survival rate 44%The chance one attempt at a company is still there after ten yearsThe p of one roll; the ticket and the ladder raise it
−NT$16.19 millionThe twenty-year net difference for the average entrantA conditional probability; the three switches decide the sign
Ticket recovery rate 0The ticket does not transfer with the assets when the company is soldPartly transferable once written into a knowledge base
Number of attemptsOnce; hard to try again after failingKnowledge output plus locations, n decided by you — provided every roll is one you can afford to lose

Parallel universes vs a timeline

Take one game and calculate it all the way. Flip a fair coin: heads, your assets are multiplied by 1.5; tails, by 0.5; a 50% chance of winning. The expected value of each round is 0.5×1.5 + 0.5×0.5 = 1.0 — no gain, no loss, and it looks like you could play forever.

First, the parallel universes. Play two rounds with NT$100 and four universes split off: win-win NT$225, win-lose NT$75, lose-win NT$75, lose-lose NT$25. The average of the four is exactly NT$100. The expected value is not lying.

Now the timeline. You live in only one of those universes. One win and one loss is 1.5×0.5 = 0.75, whichever comes first — a quarter gone. The longer you play, the closer wins and losses get to half and half, and that 0.75 keeps multiplying.

Table 3: The same game played 100 times (50 wins, 50 losses), two kinds of average

ReadingHow it is calculatedWhat is left of NT$100
Parallel universes (arithmetic mean)Each round 0.5×1.5 + 0.5×0.5 = 1.0; 100×1.0¹⁰⁰NT$100
Timeline (geometric mean)Each win-loss pair 1.5×0.5 = 0.75; 100×0.75⁵⁰About NT$0.000057

Both numbers are right; they answer two different questions. The arithmetic mean asks “if ten thousand people each play once, how much do they get back on average”; the geometric mean asks “if you alone play a hundred times, how much is left at the end.” Starting a company, running one, investing — all of them are the second question. The ergodicity problem Ole Peters wrote about in Nature Physics in 2019 is exactly the gap between these two averages.

So how much should you stake each round? In 1956, John Kelly of Bell Labs worked out this question in the Bell System Technical Journal, in what later became known as the Kelly criterion: when betting repeatedly, the fraction that makes assets grow fastest over the long run is f* = p/a − q/b, where p and q are the chances of winning and losing, b is how much each dollar gains on a win and a is how much each dollar loses on a loss. Put this game in: 0.5/0.5 − 0.5/0.5 = 0.

For a game whose expected value loses nothing, Kelly’s answer is to stake nothing at all. Any fraction you stake shrinks over repeated play; staking everything shrinks it by 13% a round on average (√0.75 ≈ 0.866). Only when the chance of winning rises to 60% does the answer change: 0.6/0.5 − 0.4/0.5 = 0.4, staking 40% each round, and the typical result after 100 rounds is about 7.5 times; with the same 60% chance but staking everything, the typical result is only about 3% of what you started with.

Back to this article: the ticket and the ladder change p. Before p has been raised, Kelly’s answer is zero; after it has been raised, you still stake only part, never everything you own. The order cannot be reversed — first make sure every roll is one you can afford to lose, then talk about how much to stake.

6. Two kinds of reader, two solutions

Part One concluded that “those with tickets don’t enter; those who enter have no ticket.” That is an allocation problem. An allocation problem has individual solutions; they are just different for two kinds of people.

For those who cannot build a ticket

Part One was clear: the cause is not inability to learn but position — a seat in execution does not produce information on “how other people decide.” The solution is to change seats: from operator to process, from floor staff to purchasing, from driving to scheduling. The salary may not change right away, but the ticket starts to build, and that compounds. Part One said the compounding loss of low pay costs more than the current loss; the reverse also holds — moving to the right seat, the compounding gain is worth more than a raise.

For those who have a ticket and do not enter

The three layers of opportunity cost are correctly counted: the NT$3.55 million bar, the value of certain income, the identity you cannot return to after failing. Not entering is rational. But the ticket expires: buyers are reassigned, engineers retire, processes move to a new generation, and a ticket held at forty is worth less than half at fifty-five. Not using it is not keeping its value; it is letting it lapse. Route B’s +NT$5.60 million assumes it is used at forty. U.S. data also show that entering in your forties or fifties is not late: founders of the top 0.1% fastest-growing new firms were on average 45 when they started.

One line for both kinds of reader: the easy misreading of Part One is to read “the expected value is negative” as “the environment is bad.” The numbers say the opposite — the parameters the environment gives are fixed; the three cells that can move are all on your side. The lottery’s −45% is set by law, the ceiling of the salary curve comes from statistics; the ticket, the ladder and the growth rate are written in nobody else’s table.

7. What this means for mid-stream supply chain companies

The 65% step: the customer stops asking — the specification names you

From 19% to 57% is the road of a company, not only of a person. At the 19% step the customer asks about price and you answer with a quote; anyone can enter and nobody makes money. Each step up corresponds to a change in the question the customer asks.

30–40% — the customer asks “will it work?”
You need an engineering team that can answer: measuring dimensions, matching power, water and air, finding compatible parts, assessing refurbishment. What you send back is not a quote but whether it will work. This is excellence; it needs nobody’s approval and can start today.

50–57% — the customer asks “what do you think?”
Your judgement itself has a price: daring to say this batch cannot ship, daring to say where the risk is, daring to advise the customer not to buy the expensive option. This is courage, and it requires the step before it to stand firm.

65% — the customer stops asking.
The specification names you. This is justice — the customer repays your value through their system — and it takes ten years with no shortcut; the shortcut is Part One’s route where 24% die in the first year.

How the amplifier is used here: write down every answer to “will it work?”, and the next customer has read it before asking, and the engineer at the second location has learned it before starting. Knowledge output is an effort-saving device for climbing the ladder, and the vehicle that carries the ticket from one person to a company.

There is only one number to watch: the annual growth rate of the surviving company. 12% is the threshold where the sign flips positive, and also where “having money” becomes “having capital” — below it, you are earning a good income; above it, the company is worth something even when you are not there. There is also only one precondition: the 12% has to grow out of your own cash flow, not be borrowed on personal guarantees.

Part One ended with this line: the expected values of the first two roads are parameters of the game; the expected value of the third is written by you.

This article only lays out “how to write it” — the ticket, the ladder, the growth rate, plus an amplifier that has existed only for the last few years. The numbers did not get better. What changed is who is reading them.

When to enter: see The Remedy, “The Window Opens Only for Those with a Floor.”

Frequently asked questions

If the expected value of starting a company is negative, why do some people succeed?

Expected value is the average of everyone who enters; different conditions give a different sign. The figures in Part One show that building ten years of industry experience before starting moves the twenty-year net difference from −NT$16.19 million to +NT$5.60 million — provided survival is 65% and annual income NT$4 million once you have the ticket, two figures that are industry estimates.

What is ergodicity, and what does it have to do with starting a company?

The average of a group is not the average of one person’s lifetime. As long as some roll can take a person straight out of the game (for example, a personal guarantee that reaches through to the individual), the more times it is repeated, the closer the result gets to zero. So starting a company means first making sure every roll is one you can afford to lose.

What is the Kelly criterion?

The Kelly criterion calculates what fraction to stake each round, when betting repeatedly, so that assets grow fastest over the long run: f = p/a − q/b (p and q are the chances of winning and losing; b and a are how much each dollar gains on a win and loses on a loss). Its answer is often smaller than intuition: in a game with a 50% chance of winning, gaining 50% on a win and losing 50% on a loss, the expected value loses nothing, yet Kelly’s fraction is 0; only when the chance of winning rises to 60% does it stake 40%. The formula comes from a paper John Kelly published in the Bell System Technical Journal* in 1956.

What is “black-ink bankruptcy”?

A company that is profitable on paper but fails because it cannot keep its cash flowing. It can happen when growth outpaces the “self-financeable growth rate” the company’s own cash flow can sustain.

At what age is starting a company most likely to succeed?

U.S. Census Bureau data show that the chance of entrepreneurial success generally rises with age up to around 60; a 50-year-old founder is about 1.8 times as likely as a 30-year-old to build a top-growth company. Founders of the top 0.1% fastest-growing new firms were on average 45 when they started; the longer and closer their time in the same industry, the higher the success rate. In this series’ terms: the ticket only builds while you stay in the industry, and it expires if you are away too long.

Can experience be written down and passed on?

Partly. Once tacit knowledge is written into documents, it moves from one person’s ability to a company’s asset; the part that cannot be written down, tied to the shop floor, is the part nobody else can copy.

Road to NT$100 Million series: Part One, “The Odds and Costs of Three Roads” | Part Two, “You Write the Sign” | The Remedy, “The Window Opens Only for Those with a Floor” | Part Four, “Margin Is the Price of Trust”

References

[1] JIAHON Think Tank (2026), Choosing a Road to NT$100 Million — Lottery, Salary, Startup: The Odds and Costs of Three Roads (Part One of this article). All figures in this article are taken from it; no new estimates were made.
[2] JIAHON Think Tank (2026), “Four Pillars, Build Your Own Two First: Truth → Excellence → Courage → Justice”.
[3] Ministry of Economic Affairs, White Paper on Small and Medium Enterprises, annex tables 11, 13, 16; tables 5-1-4, 5-1-5.
[4] Ministry of Finance, 2025 (ROC 114) Industry Income and Profit Standards for Profit-Seeking Enterprises (1,283 categories; net margin = 0.348 × gross margin + 1.51, r = 0.897).
[5] Ministry of Finance, national business (tax) registration datasets (suspended / not operating, other than suspended); DGBAS, Census of Industry and Services (2016, 2021).
[6] CTBC, BCG (2026), Taiwan Ultra-High-Net-Worth Wealth Insights Report; Taiwan Lottery, 2025 (ROC 114) jackpot statistics.
[7] The three sets of figures in Figure 1 are each taken from the separate models in Sections 6, 7, 8 and 10 of Part One, with no combined calculation; the −NT$15.69 million in the growth scenario is the baseline of Part One’s eighth-version (62×) model.
[8] Ole Peters (2019), “The ergodicity problem in economics,” Nature Physics 15(12): 1216–1221. https://doi.org/10.1038/s41567-019-0732-0
[9] Warren Buffett (15 October 1998), lecture at the University of Florida business school (on Long-Term Capital Management).
[10] Neil C. Churchill, John W. Mullins (2001), “How Fast Can Your Company Afford to Grow?,” Harvard Business Review, May 2001 (R0105K).
[11] Michael Polanyi (1966), The Tacit Dimension, Doubleday.
[12] Ikujiro Nonaka, Hirotaka Takeuchi (1995), The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press.
[13] Nassim Nicholas Taleb (2018), Skin in the Game: Hidden Asymmetries in Daily Life, Random House, chapter 19, “The Logic of Risk Taking.”
[14] Warren Buffett (1984), “The Superinvestors of Graham-and-Doddsville,” Hermes, Columbia Business School. https://business.columbia.edu/insights/chazen-global-insights/superinvestors-graham-and-doddsville
[15] Pierre Azoulay, Benjamin F. Jones, J. Daniel Kim, Javier Miranda (2020), “Age and High-Growth Entrepreneurship,” American Economic Review: Insights 2(1): 65–82. https://doi.org/10.1257/aeri.20180582
[16] John L. Kelly Jr. (1956), “A New Interpretation of Information Rate,” Bell System Technical Journal 35(4): 917–926. https://doi.org/10.1002/j.1538-7305.1956.tb03809.x

Note

This is an opinion article. All figures are taken from Part One of this series and the public statistics it lists; Part One states the data grade of each item (hard data, calibrated, industry estimate, assumption), and this article keeps that grading and adds no new assumptions. This article does not constitute investment, entrepreneurship or legal advice.

Corrections

Revised 2 October 2026: eight external references on ergodicity, the self-financeable growth rate and tacit knowledge were added, along with the section “Before the switches: the floor” and the FAQ, and the following were corrected:
① the net difference quoted from Part One in Section 1 originally read NT$15.54–15.98 million; corrected to NT$15.65–16.02 million;
② “three times the gap” in Section 5 was corrected to “nearly three times” (80 ÷ 30 = 2.67).
Further corrections the same day: ③ Section 1 originally said these numbers “were measured, not estimated”; corrected to say the survival rates and revenue brackets were measured, while the amounts in the net difference contain assumptions; ④ “Before the switches: the floor” originally said Part One “measured” the recovery rate; corrected to “estimated” (the recovery rate is an industry estimate); ⑤ the timing of Buffett’s remarks on Long-Term Capital Management was corrected from “when it collapsed” to “when it came close to collapse”; ⑥ the FAQ “At what age is starting a company most likely to succeed?” originally answered “45 on average”; rewritten to the study’s actual finding: success rises with age to around 60, and 45 is the average founding age of founders of top-growth new firms.

License

This work is published by JIAHON Group. All rights reserved. Quotation and redistribution are permitted with proper attribution and a link to jiahon.com.tw. Commercial use—including but not limited to paid courses, reports, or consulting services incorporating this content or its frameworks (including “three switches and one amplifier” and “the margin ladder as a projection of the pillar order”)—requires prior written authorization from JIAHON. For commercial licensing enquiries, please contact us through the website.

Back to top