The idea

A 10-bagger is a term introduced by investor Peter Lynch to describe an investment—typically a stock—that appreciates tenfold from its original purchase price.

When people talk about 10-baggers today, in markets very different from those of 40 or 50 years ago, a common claim keeps resurfacing:

The strongest predictor of 10-bagger returns is fast and consistent revenue growth before profitability.

The logic usually follows a familiar path:

1. Peter Lynch introduced the concept of the 10-bagger.

2. Markets have changed:

   • many companies grow first and monetize later

   • especially in technology

3. Iconic examples are often cited:

   • Tesla

   • Amazon

   • Shopify

4. All of them:

   • were unprofitable for years

   • but showed extremely high revenue CAGR

5. Quantitative observations reinforce the narrative:

   • more than 300 “true” 10-baggers over the last 20 years

   • revenue growth is typically described as:

     • well above 20% during early acceleration phases

     • often exceeding 30% in the years that follow

   • compared with roughly 5% for the S&P 500

The idea is seductive.

Study the greatest winners of the last 20 years.
Extract what they had in common.
Replicate it.

The data looks compelling.
The examples are real.
The logic feels intuitive.

The problem is not that the idea is wrong.

The problem is what happens when it is applied in real time.

1. Why these examples are so convincing

Looking at a handful of iconic companies reinforces a powerful implicit message:

• Look how far they ran before becoming profitable.

• The market anticipated value.

• Profitability came later, almost as a formality.

The implicit conclusion is even stronger:

If 10-baggers belong to a structurally different regime,
and that regime can be identified early,
then exceptional returns should be repeatable.

From a narrative standpoint, everything fits neatly.

2. Where the problem really starts (ex-post, not theoretical)

This is where we enter the core territory of When Ideas Meet Reality.

Statements that link unusually high revenue growth to the “start” of a major price run often sound precise—but they hide a critical flaw.

Stripped of narrative framing, the logic is essentially this:

Looking backward at companies that eventually became 10-baggers,
in the year we now label as the start of the run,
revenue growth was already unusually strong.

The issue is obvious.

In real time:

• you do not know that this is “the year the run begins”

• you only know it after the stock has already gone 5×, 10×, or more

This is not very different from saying:

“All major trends start from a well-defined low.”

Always true.
Almost never useful.

This is pure selection bias:

• start from the winners

• look backward

• extract a common trait

• present it as a predictor

It is not a lie.
But it is not a standalone, replicable strategy.

In other words, the dataset is conditioned by the final outcome.

We observe:

• only companies we already know became 10-baggers

• and then examine what they had in common beforehand

This produces ex-post correlation, not necessarily:

• an ex-ante operational rule

• or a filter usable in real time

3. The implicit assumptions no one states

The argument only works if several unstated assumptions hold true:

1. Revenue growth ≈ value creation
   (not always true)

2. Markets reward growth even when it is not yet monetizable
   (highly dependent on interest rates, liquidity, and narrative regimes)

3. Growth is sustainable
   (in most cases, it is not)

4. Dilution, unit economics, and capital intensity are secondary
   (they are not)

5. The path does not matter, only the final outcome
   (true ex-post, false ex-ante)

These assumptions are not irrational—but they are fragile.

The most dangerous confusion among them is this:

Revenue growth is not the same as value creation.

It is true that:

• many extraordinary companies grow aggressively before profitability

• stock prices often accelerate around the first sustainable profits

But it is equally true that:

For every Amazon or Tesla, there are dozens of companies that followed the same early script—and failed.

Why?

Because revenue growth can be:

• purchased through discounts and aggressive marketing

• structurally unscalable

• dependent on continuous external capital

• cyclical rather than durable

Markets are full of companies that:

• grow revenues rapidly

• reach profitability for a quarter or a year

• return to cash burn

• dilute shareholders

• and eventually collapse

The problem is not profit now versus profit later.

The problem is the quality of the growth.

4. What revenue-centric frameworks usually omit

Most revenue-based narratives ignore at least four decisive filters:

Unit economics

• Do margins improve with scale?

• Does each incremental customer create more value or less?

Capital intensity

• Can growth progressively self-fund?

• Or does it require constant capital injections?

Durability

• Is the growth defensible?

• Or can a well-funded competitor erase it?

Timing and path dependency

• Many 10-baggers reward only very early entrants

• When “the story becomes obvious,” returns are often already compressed

None of this is captured by revenue CAGR alone.

Yet these factors determine whether growth compounds into value—or collapses under its own weight.

5. The real conceptual tension

The critical point is not:

“Revenue growth does not matter.”

The critical point is this:

• all 10-baggers grow rapidly

• but most companies that grow rapidly do not become 10-baggers

Ideas live in spreadsheets.
Investments live in time.

Time introduces:

• uncertainty

• competition

• capital constraints

• human behavior

• shifting incentives

A framework that looks flawless when applied to finished success stories
can break down completely when applied to unfinished ones.

This is where many elegant ideas fail.

Not because they are wrong—
but because reality is messier than backtests.

6. The honest conclusion

Stripped of marketing, the correct statement is simple:

Rapid and consistent revenue growth is a necessary—but not sufficient—condition for exceptional returns.

Without:

• scalable margins

• disciplined capital allocation

• durable competitive advantages

the probability of value destruction remains far higher than that of finding a 10-bagger.

This sentence does not sell newsletters.

But it is true.

7. How to use this insight without falling into the trap

Revenue growth should not be treated as a strategy.

It is better understood as:

• a starting condition

• a narrative signal

• a way to avoid stagnation

Markets love to tell stories about winners.

Investing, instead, is about surviving errors, not just recognizing success in hindsight.

When ideas meet reality, the difference is not growth.

It is whether a business can survive its own expansion.

The end of a year is not a time for certainties, but for traces.

Markets do not close the year by leaving instructions.
They leave partial, often contradictory signals, reflecting expectations, constraints, and priorities that are beginning to surface — not necessarily those that will ultimately prevail across global markets, from US equities to currencies and capital flows.

2025 closed with one element that is difficult to ignore: a meaningful underperformance of US equities relative to the rest of the world.

The most frequently cited figure is simple: on an annual basis, the US equity market underperformed the rest of the world by roughly 10%.

From there, a series of conclusions is often presented as almost self-evident: a weaker dollar ahead, rotation toward ex-US markets, a renewed role for commodities, and the need to rethink capital allocation.

It is an elegant narrative.

But before turning it into a decision, it is worth pausing and doing what every investor should do first: asking questions.

The questions that are often not asked

The first question is also the most uncomfortable one:

who is driving whom?

Is the underperformance of US equities weakening the dollar?
Or is it the dollar — when it weakens — that reduces the relative advantage of US markets versus the rest of the world?

The difference is not academic.

In the first case, equity markets would be reflecting a weakening real economy that later transmits to the currency.
In the second, it is monetary and financial dynamics that redirect capital flows, with subsequent effects on equity markets.

If today the dollar remains structurally strong against many currencies (the euro included), on what basis should we conclude that US equity underperformance marks the beginning of a new era, rather than a temporary divergence?

Taken in isolation, this data point does not tell a complete story.
But ignoring it altogether would be equally short-sighted.

The question is not whether we are witnessing an “epochal shift.”
The question is what the market is beginning to challenge — and what it is still implicitly supporting.

1. US vs. the rest of the world: what does this chart really tell us?

The chart below shows the relative performance of the S&P 500 versus world equities ex-US, observed year by year since 2009.

Each grey line represents a historical year.
The black line represents the year just concluded.

Toward the end of the year, the black line sits around –10% to –12%, a zone that historically is not frequent and rarely goes unnoticed.

This does not automatically mean:

• that the US market is “finished”

• that the rest of the world is about to outperform for years

• that we are entering a new structural regime

It means something simpler — and more interesting:

the market has stopped treating the United States as the default answer.

When a relative divergence reaches this magnitude, it ceases to look like a tactical rotation and begins to behave like a message that deserves interpretation, not dismissal.

2. The central node: a strong dollar or strong US markets?

Here emerges the key question, often taken for granted:

was it a strong dollar that sustained US outperformance over the past years,
or was it US outperformance that sustained the dollar?

The distinction is crucial.

Historically, two broad regimes can be observed:

• Strong dollar → US outperformance
  (global capital seeking safety, liquidity, and depth)

• Weak dollar → rest of the world outperformance
  (rotation toward real assets, cyclical markets, peripheral currencies)

In recent years we have seen:

• a broadly strong dollar (especially versus the euro)

• dominant US equity markets

• highly concentrated global capital flows

What makes 2025 interesting is that this relationship has begun to fray.
The dollar has not collapsed, yet US outperformance has weakened.

This divergence is not an answer.
It is an open question the market is now asking.

3. Commodities: real demand or liquidity effect?

Another frequently repeated narrative is the following:

“If monetary policy becomes more accommodative, commodities will rise.”

Reality is more complex.

In theory, commodity prices reflect:

• real demand

• income growth

• rising economic activity

But the most recent cycle has shown something different.

In Europe, inflation was not driven by domestic demand, but by:

• energy shocks

• rising production costs

• price increases that were forced rather than expansionary

In the United States, inflation was more consistent with:

• fiscal stimulus

• prolonged monetary easing

• genuine real growth

This creates an anomalous environment:

• global monetary easing

• strong financial markets

• real economies that remain misaligned across regions

As a result, the logical chain
rate cuts → real growth → commodity boom
is far less automatic than often assumed.

Over the past fifteen years, financial markets have repeatedly risen much faster than the real economy, without global demand following in proportional fashion.

4. Why today’s picture is ambiguous, not directional

Another rarely articulated point concerns why monetary policy becomes more accommodative.

Rate cuts can be:

• expansionary, when they accompany a temporary slowdown

• defensive, when they aim to cushion a genuine recession

• financial, when they serve to stabilise a highly indebted system

Each of these scenarios has very different implications for:

• the dollar

• equity markets

• real assets

• future returns

Without clarifying the reason for easing, there is a real risk of confusing nominal signals with structural change.

5. US valuations: not a collapse, but a lower ceiling

Another element the market seems to have priced quietly concerns valuations.

Aggregate indicators such as the ratio of total market capitalisation to GDP (often referred to as the “Buffett Indicator”) remain historically elevated, around 2× GDP.

This does not imply:

• an imminent crash

• a repeat of the year 2000

It implies something more restrained:

• less room for multiple expansion

• returns increasingly dependent on real earnings growth

• greater sensitivity to shocks

In such environments, markets can still rise.
But they must earn it, quarter after quarter.

6. The real question for investors today

At this point, the issue is no longer macro.
It is decisional.

- Does it still make sense to treat US markets as the sole reference point for allocation?
- Or is the market beginning to suggest looking elsewhere, without dogma and without haste?

And above all:

- Is the signal we are observing structural or transitory?

For now, the market has not provided a definitive answer.
But it has clearly indicated that some of the certainties of the past decade are no longer automatic.

Conclusion: what to do — and what not to do

This is where the most important part begins.

When the picture is unclear, the worst mistakes are often:

• chasing a narrative because it “sounds convincing”;

• acting simply because “everyone is talking about it”;

• outsourcing judgment to those who speak with greater confidence.

Understanding why a relationship should work matters more than knowing that it worked in the past.

Not understanding is not a fault.
Acting without understanding is.

Sometimes, the most rational decision is not to reallocate capital, but to wait until causal relationships become clearer.

2025 does not hand us a new truth.
It hands us friction.

And in markets, it is often within frictions — not ruptures — that the best decisions take shape.

For now, the message is not “change everything.”
It is to stop taking everything for granted.

And to watch how capital actually reallocates when uncertainty outweighs conviction.

But when I have to make a real decision, I don’t feel more confident than before.”

This experience is far more common than it seems.

It shows up in people who study seriously.

Over time, knowledge accumulates.
Concepts pile up.

Each one makes sense on its own.
Each one feels useful.

But when decisions are required, those concepts don’t line up.
They point in different directions.

What was meant to create clarity often creates friction instead.

As a result, decisions become harder, not easier.

Understanding grows.
Confidence doesn’t.

Action gets delayed.
Or it becomes reactive.

This is where many investors draw the wrong conclusion.

They assume the issue is incomplete knowledge.
That one more concept, one more model, one more refinement will finally resolve the tension.

So they keep studying.
And the gap quietly widens.

Investors often recognize the right concepts.
- Return on capital
- Cash flows
- Moats
- Valuation

Traders experience the same tension in a different form.
Signals, setups, indicators, risk rules — each reasonable on its own, yet rarely aligned when a decision is required.

The issue isn’t their relevance.
It’s how they coexist — or fail to — inside a single decision.

This publication exists to examine that gap.

Not by adding more tools.
Not by offering shortcuts.
And not by pretending certainty can be engineered.

The focus here is on how decisions are formed before they turn into action —
and how small distortions in reasoning compound long before capital does.

As Charlie Munger once put it:

“You have to master the big ideas and commit yourself to slowly improving your judgment, as we all had to do.”

Understanding comes before autonomy.
Autonomy comes before action

Over time, I’ve learned that most investment mistakes don’t come from missing information, but from unclear thinking before action.

This page is where I try to keep that thinking organized.

The posts collected here are grouped by theme, not by chronology. Each section reflects a different angle of the same process: how ideas are formed, how markets react, and what happens when those ideas are tested against reality.

This page is updated over time as new posts are published.

Foundations / Capital Thinking

How investors learn to think about capital, incentives, and long-term outcomes — beyond formulas and metrics.

What the Market Revealed

Observations on what market behavior reveals about expectations, risk, and capital

When Ideas Meet Reality

What happens when ideas are exposed to real constraints, incentives, and time.

The problem is not determining whether AI is real.

The problem is understanding what kind of investment we are actually buying when we say “AI.”

On one side, there are those who argue we are facing an unsustainable bubble, fueled by massive investments and companies like OpenAI burning cash at unprecedented rates.
On the other, there are those who claim the dot-com comparison is misleading: today’s big tech companies have solid balance sheets, real profits, and tangible infrastructure.

The field is split in half.

And this is where many investors ask the obvious question:
Which side should I be on? Should I invest in this business or not?

That is the wrong question.

It is a question driven by timing anxiety, not by an understanding of the business itself.
If you are looking for the next 10x, it may not be an unreasonable question.
But if you are looking for predictability and investment rationality, asking it will likely leave you paralyzed—never knowing whether entering or staying out was the right decision.

The real question is different:
What do I want from this investment? What return am I expecting? And is it predictable?

Let’s take a concrete case: SoftBank and ARM.
Who are they—and why are they often treated as proxies for OpenAI?

Over the past 12–18 months, SoftBank (listed in Tokyo) has explicitly tied its future to AI.
Its CEO has described SoftBank as “the orchestrator of the AI ecosystem,” publicly stating that “AI is the new secular cycle” and positioning OpenAI as the conceptual reference point.

In practice, SoftBank is betting on the infrastructure that makes OpenAI—and AI more broadly—possible.
Buying SoftBank means buying a concentrated wager on AI.

SoftBank controls roughly 90% of ARM Holdings plc (listed on Nasdaq as ARM).

So what does ARM actually do?

ARM does not manufacture chips.
ARM licenses architectures.

Every ARM-based chip pays:

· a licensing fee

· a per-unit royalty

ARM is embedded in:

· smartphones (Apple, Qualcomm)

· edge computing

· embedded systems

· and increasingly, data centers and AI inference

Translation for technically hostile minds (like mine…):
ARM sells the base language used to design many chips and collects a royalty every time those chips are sold. Full stop.

ARM is not interesting here as a stock.
It is interesting as a stress test for an investment thesis built on narrative.

SoftBank has invested—directly or indirectly—in:

· data centers

· chip design

· cloud infrastructure

· AI-specific hardware companies

Not individual AI models, but what allows those models to run.

It has also sold stakes in Nvidia and Oracle—not to exit AI, but to further concentrate its exposure on what it considers core AI infrastructure.

This is the opposite of diversification.

ARM is valuable because:

· it is a de facto standard

· it lives off royalties

· it has theoretically very high margins

· it is classic “picks and shovels”

And indeed:

· ARM’s valuation runs into the tens of billions

· its weight in SoftBank’s net asset value is dominant

So far, so good.

Over the past two years, analyst consensus on ARM has been overwhelmingly positive (buy, outperform, and so on).
Its cash flow statement shows positive free cash flow, positive operating cash flow, and very high gross margins.

But there is a major crack in the story.

RISC-V.

· open-source

· royalty-free

· strongly supported in China

· increasingly used for custom AI accelerators

If:

· AI becomes a commodity

· hardware becomes more customized

· hyperscalers seek to reduce lock-in

ARM does not lose volume.
It loses pricing power.

That is lethal for a “royalty compounder” thesis.

A reasonable objection follows:
AI is now indispensable. Every major tech company is involved. Analysts broadly agree. ARM may be expensive, but its fundamentals are strong—surely not everyone investing in ARM is irrational.

A fair question.
But once again, it is not the right one.

The relevant question for a value investor is:
Can I evaluate this business autonomously, or am I following a narrative?
Can I identify and assess the real risks involved?

The answer does not require being a financial guru, holding an MBA, or having an exceptional IQ.

Today’s price discounts future rents that have not been demonstrated, because it assumes:

· dominance

· lock-in

· regulatory stability

· absence of viable alternatives

These are assumptions, not facts.

In November 2025, something narrative-relevant happens: Alphabet introduces Gemini 3.0.

The message to the market is simple:
OpenAI is not untouchable.

This does not immediately affect SoftBank’s revenues or near-term fundamentals.
But it does affect:

· the leadership narrative

· the idea of a single AI winner

· the justification for the most aggressive valuations

And SoftBank, by construction, is hyper-narrative.

The result: SoftBank and ARM sell off.
Goldman Sachs issues a downgrade in mid-December.

None of this means that ARM is a bad investment.
A fund could buy ARM today and be up 30% in 6–12 months. That is entirely possible.

But that is not the point.

The point is that a value investor asks simpler questions:
Is this business rational?
Is it predictable?
Do I truly understand it?

The straightforward answer is that its value depends on too many simultaneous conditions:

1. that it remains a standard

2. that it is not bypassed

3. that open-source alternatives fail to gain traction

4. that governments do not interfere (China, in particular)

5. that AI does not alter chip architectures

It is a castle of ifs.

A value investor immediately senses this:
This is not a certainty that accumulates.
It is a narrative that must be defended.

Which explains why:
great technology ≠ great investment

This does not mean a value investor should only invest in simple, low-tech businesses.
It means something else.

I must be able to explain:

• where value comes from

• why it should persist

• and what could break it

When evaluating an investment—before reading a balance sheet or a report—I mentally run through a few core pillars of disciplined investing:

· Narrative vs fundamentals
Is the story powerful, and does it actually match the business?

· Declared moat vs understandable moat
If the moat only exists after 20 minutes of technical explanation, is it really a moat?

· Predictability vs sophistication
I don’t need the most brilliant business; I need the most legible one.

· Implicit, invisible risks
If I can’t list them without getting a headache, I’m not controlling risk.

That is a real competitive advantage.
And it brings us back to the initial question:
What do I want from my investment? Can I predict it? And how much risk am I taking to get there?