Value investing

Business-owner mentality

A stock is fractional ownership of a real business — not a ticker, not a chart pattern, not a piece of paper to trade. Once you adopt the owner framing, decisions about "should I hold this" become decisions about "do I want to own this business for five years". The set of acceptable answers shrinks fast.

The framing changes specific decisions in four concrete ways:

• Volatility. The price of a private business doesn't tick every second; an owner thinks about value over years, not minutes. Applied to public stocks: stop watching the price more than once a quarter unless something specific has happened to the business.
• News cycles. The owner asks "does this story change the fundamentals of my business?" not "what does the market think of this story?" Most news is the second; only a fraction is the first. The owner thinks about business impact, not headlines.
• Capital-allocation announcements. A buyback is the company buying back YOUR ownership. A dividend is the company paying YOU cash. An acquisition is the company spending YOUR balance sheet. Each gets evaluated through that lens — would the owner of a private business approve this trade at this price?
• Long-term holding. Owners don't try to time their entry and exit; they pick businesses they'd be content holding for a decade and let compounding do the work. Public-stock holders who match that horizon access the same compounding.

Being a minority shareholder: a 0.0001% owner of a public company has no operational control — can't fire the CEO, can't veto an acquisition, can't redirect strategy. The owner framing has to be philosophical (how you think) rather than operational (what you do). The exception is concentrated ownership (5%+ stakes) where activist mechanics open up — a separate game with separate skills, not what this workshop is about.

Intrinsic value discipline

The market gives you a price every minute; intrinsic value is what you do separately, on your own, away from the screen. Holding both numbers in mind without letting one dominate the other is the discipline. The Ginza-building thought experiment from the slides ("if a building were selling at $100K, you wouldn't conclude it's terrible — you'd conclude it deserves investigation") is the mindset in one image.

Margin of safety

Buy a company at a price that is sufficiently below intrinsic value so that you can be wrong and still come out OK. The margin (i.e., the difference between what you paid and what it's worth) is what protects against the unknown unknowns. Higher margin means lower required precision in the valuation; you can afford to be roughly right rather than precisely right.

Worked example. Estimate a business is worth $100/share. Market price is $60. Margin of safety is 40%. If the valuation is wrong by 50% (the business is actually worth $50), the position still breaks even at $60 — a painful 17% loss. However, if the purchase price was $90 (10% margin of safety), the same 50% valuation miss is a catastrophic 44% loss. Same analytical mistake; very different financial outcome — driven entirely by the entry margin.

How big a margin to require depends on four factors:

• Business quality. A high-quality compounder (durable moat, capital-light, strong management) needs less margin — time is on the buyer's side. A low-quality cyclical needs more — the business itself can deteriorate while you wait.
• Conviction. Strong, well-supported thesis → can accept smaller margin. Weak or speculative thesis → demand much larger margin to compensate for thesis-error risk.
• Cyclical position. Buying at the top of an industry's cycle requires larger margin (next-year earnings will likely be lower than today's). Buying at the bottom requires less (next-year earnings will likely be higher).
• Balance-sheet risk. A leveraged business deserves a much larger margin than a debt-free one — the downside includes equity wipeout, not just price decline.

The interaction with sizing: margin of safety is one of the four inputs to position size (see Research → Stage 4). Larger margin of safety on a high-quality business with high conviction is what justifies the rare 10% position. Smaller margin and weaker conviction caps the position at 2–3% no matter how attractive it looks.

1. Asymmetric thinking

The compounding asymmetry is the foundation. A -50% loss requires +100% to break even. A -75% loss requires +300%. A -90% loss requires +900%. The downside is bounded at -100% (the position can't go past zero), but the recovery required to undo it scales non-linearly.

The corollary is sequence-of-returns risk: when losses happen matters as much as how big they are. A -30% loss at age 30 happens against a small portfolio with decades of compounding ahead — recoverable. The same -30% loss at age 65 happens against a peak portfolio that's already funding retirement — much harder to recover from, because you're now selling assets at the bottom to cover living expenses. Treat late-career drawdowns as fundamentally more dangerous than early-career ones; size your risk-taking accordingly across the lifecycle.

2. The kill-thesis discipline

Write the investment's obituary before buying it. The prompt: "three years from today, this position lost 70%. What happened?" Write the answer in specific terms — what changed about the business, what catalyst triggered it, what the financials would have looked like as it unfolded.

This forces confrontation with the bear case while you're still emotionally detached. The investor who skips the kill-thesis often discovers, two years later mid-drawdown, that they never had a clear view of how the investment could fail — which means they have no way to distinguish "this is the thesis breaking" from "this is normal volatility." Without that distinction, holding-vs-selling decisions become emotional rather than analytical. Full memo structure on Research → Stage 3.

3. Position sizing as risk control

The Kelly criterion — bet a fraction of capital proportional to edge divided by odds — gives the mathematically optimal sizing if probabilities and payoffs are known precisely. Real investing has none of that precision, so practitioners use fractional Kelly (typically half-Kelly or quarter-Kelly) to compensate for over-confidence in the inputs. Even half-Kelly produces position sizes that feel uncomfortably large to most amateurs — and uncomfortably small to most professionals.

The practical rules of thumb:

• Hard maximum on single position: 10–15% of the active sleeve, even for highest-conviction ideas. The asymmetry argument from above: a 25% position falling 50% takes the whole portfolio down 12.5% — recoverable. A 25% position going to zero takes it down 25% — much harder.
• Typical alpha-generator size: 3–5% of active sleeve. Enough that a 100% return moves the portfolio meaningfully; small enough that being completely wrong doesn't kill the year.
• Cluster risk: Don't let three "different" positions all be levered to the same underlying factor (e.g., three different semiconductor names — all bet on the same AI cycle). The portfolio's real concentration is whatever the largest factor exposure is, not the number of tickers.

4. Survival > optimization

The single most expensive mistake an investor can make is the one that takes them out of the game. Compounding only works if there's still capital to compound; an investor who blows up at year 5 of a 35-year horizon doesn't get to participate in the compounding that would have done most of the work in years 25–35.

Applied: prefer the slow-but-survivable strategy over the fast-but-fragile one. The 12%-per-year strategy with no leverage and no concentration risk beats the 25%-per-year strategy that has a 15% chance of wiping out in any given year. Run the math over 30 years and the slow strategy compounds to roughly $30 per dollar invested; the fast strategy has an expected terminal value below the slow one because the wipeout scenarios dominate.

This is the Buffett rule restated: "don't lose money" isn't moralistic; it's mathematical. Survival of capital is a prerequisite for compounding, and compounding is the entire game.

Layer 1 — Fundamental

What is this business worth? Form your own independent view, before looking at the price. Walk through earnings power, asset value, growth value (the three Greenwald lenses on the Frameworks page). Decide what you'd pay if the market didn't exist.

Layer 2 — Sentiment

What does the market currently believe about this business? Not just the price — the story that produces the price. Read sell-side notes, scan financial media, look at the short-interest, listen to recent earnings-call Q&A. Triangulate consensus.

The gap is the opportunity

If your view matches consensus, there's no edge — you'll earn the market return. The opportunity is when you can articulate why the market's belief is wrong, AND you can articulate what changes that. Variant perception is the term of art (covered in Research Stage 3); the mindset is independent thinking that survives crowd pressure.

Patience as a structural advantage

Most professional fund managers are graded quarterly. Three bad quarters in a row and clients start asking questions; four bad quarters and capital starts leaving. The structure forces a short-term focus regardless of the manager's personal preference — a manager who knows a thesis will pay off in three years still cannot survive three years of underperformance to find out.

The retail investor faces no such pressure. There is no quarterly review with clients, no risk committee, no benchmark to track against month by month. The ability to wait three years for a thesis to play out — through periods when the position looks wrong, the price drifts sideways, the news is uniformly negative — is a structural edge that no professional can match. It is also the edge most retail investors throw away by checking prices daily and feeling pressure to "do something" when nothing needs doing.

The discipline: define the holding period at purchase, not after. A compounder thesis is a 5–10 year position; an alpha-generator thesis is a 1–3 year position; a special-situation thesis (spinoff, post-bankruptcy) is a 6–18 month position. Match expectations to time horizon; ignore short-term price action that doesn't change the thesis.

Contrarianism as temperament

A non-consensus view, by definition, is one most other people disagree with. Holding it through the period when you look wrong — and you will look wrong, often for most of the holding period — requires temperament that doesn't depend on external validation. The investor who needs the market to agree with them before they can hold a position has no edge; their thinking is downstream of consensus, not independent of it.

Contrarianism is most often misunderstood as "betting against the crowd." It isn't. It is the willingness to hold a different view when the analysis supports it, regardless of which side of the consensus that view lands on. Sometimes the contrarian position is bullish when the crowd is bearish (the META 2022 trade). Sometimes it is bearish when the crowd is bullish (sitting out the 2021 SPAC bubble). Sometimes it is exactly the consensus position because the consensus is right (most of the time, in efficient markets). The temperament is independence, not opposition.

The calibration: contrarian vs. stubborn

The failure mode of contrarianism is stubbornness — refusing to update the view when the facts change. The two look identical from the outside (both involve holding a position the market disagrees with) but they are opposite stances inside the investor's head.

• Contrarian: "My thesis is X. The market believes not-X. Here's why I think the market is wrong: [specific mechanism]. If the market is right and I'm wrong, here's what I would see: [specific kill-thesis signals]."
• Stubborn: "My thesis is X. The market believes not-X. I'm right because I'm always right / I've done more work than the market / I just feel it."

The calibration test: when the kill-thesis signals appear, does the investor update or rationalise? A real contrarian sells when the bear case starts materialising, even at a loss, because they were positioned on probabilities and the probabilities changed. A stubborn holder finds reasons to dismiss each signal and rides the thesis to zero. The difference is invisible until the test arrives, which is why the kill-thesis discipline above is load-bearing — it forces the contrarian's exit criteria into writing before the emotional cost of selling makes them hard to invoke.

Anchoring on purchase price

Mechanism. The brain treats the original purchase price as a special, meaningful number — "I'm not selling until it gets back to $100" — even though the price you happened to pay has zero relevance to whether the position is attractive going forward.

Where it shows up. Refusing to sell losing positions because doing so would "lock in the loss." Refusing to buy more of a position you still believe in because "I already bought at a lower price." Both decisions are made against your past purchase price, not against the current opportunity set.

Pre-commitment. When reviewing any position, ask: "if I had no current position and the cash equivalent, would I buy this at today's price?" If yes, hold or add. If no, sell. The purchase price doesn't enter the decision.

Endowment effect

Mechanism. People overvalue things they already own simply because they own them. In investing, this shows up as defending the existing portfolio against any change — even when the change is clearly beneficial.

Where it shows up. Holding a stale position because selling and re-buying somewhere new feels like more work, even when the new opportunity is obviously better. Resisting portfolio rebalances. Filtering new ideas through a "do I really want to displace what I already own" lens that biases toward inaction.

Pre-commitment. Quarterly portfolio review with a blank-slate frame: "if I had this cash today, would I construct this exact portfolio?" Anything you wouldn't buy fresh is a candidate to sell.

Sunk cost fallacy

Mechanism. Past investment of time or money (research hours, capital deployed) creates a felt obligation to "see it through" even when the forward analysis says exit. The hours already spent feel wasted if the position is sold; in reality they're sunk regardless.

Where it shows up. Holding a position you've researched extensively, past the point where the thesis has broken, because exiting would feel like wasting the research. Continuing to add to a thesis-broken position to "average down" — converting one mistake into a bigger one.

Pre-commitment. Apply the kill-thesis from purchase rigorously. When the kill-thesis triggers, exit — the research hours are already spent regardless of what you do next; the only question is whether the next dollar deployed has positive expected value.

Narrative fallacy

Mechanism. The human brain prefers coherent stories to messy probabilities. A neat narrative ("this company will dominate AI infrastructure") feels more confident than a messy probabilistic statement ("there's maybe a 30% chance this company captures 15% of a market that itself might be 2-5x today's size in 5 years"). The neat story is usually wrong because it papers over the uncertainty.

Where it shows up. Falling for compelling pitches at investor conferences. Buying after reading one impressive Substack post. Anchoring on management's vision presentation without checking whether the unit economics support it. Anything that feels like "this story makes too much sense for me NOT to buy" is probably narrative fallacy.

Pre-commitment. Force probabilities into every thesis. Instead of "this will work because X," write "this has roughly Y% chance of working through mechanism X; here are the specific things that would update Y up or down." If you can't translate the story into a probability, the thesis isn't ready.

Confirmation bias (especially in the kill-thesis)

Mechanism. Once a position is owned, the brain selectively notices information that supports the original thesis and discounts information that contradicts it. The kill-thesis exercise from earlier is specifically designed to interrupt this — but only if it's used honestly.

Where it shows up. Reading bullish sell-side reports and ignoring bearish ones. Dismissing critical questions from skeptical friends as "they don't understand the thesis." Reinterpreting bad news as good news ("the stock is cheap now"). Selectively quoting earnings-call lines that support the view.

Pre-commitment. Maintain a written kill-thesis from purchase. When new information arrives, ask explicitly: "does this support, contradict, or trigger my kill-thesis?" Force the categorisation. If the kill-thesis triggers — even partially — re-engage the analysis from a sceptical stance, not a defensive one.

Action bias during drawdowns

Mechanism. When a position is down sharply, the brain treats "do something" as inherently better than "do nothing," even when nothing has changed in the business. Action feels like control; inaction feels like helplessness. The pressure to act is strongest exactly when discipline is most needed.

Where it shows up. Selling at the bottom because "at least I'm in control of something." Doubling down on a thesis-broken position because "at these prices the math is even better." Reacting to every news headline during a drawdown with a trade. Switching strategies entirely because the current one isn't working in the current week.

Pre-commitment. The quarterly review cadence (see Research → Stage 4) creates a structural delay between impulse and action. Position decisions get made at quarter-end based on whether the thesis has changed, not in the middle of a drawdown week. Inside the quarter: the only action is "do nothing unless the kill-thesis has triggered." Make this a written rule, not a felt one.

FOMO and panic — the parent emotions

The six traps above are specific failure modes of two underlying emotions: fear of missing out (FOMO) when a position runs up, and panic when a position drops. Both are natural emotional responses — feeling them is not the problem. The operating principle the workshop holds to: FOMO and fear are OK; feel them, don't obey them.

The emotions exist whether the investor wants them to or not. Pretending they don't is denial, and denial breeds bigger errors than acknowledgement does. What separates the disciplined investor from the undisciplined one is the gap between feeling the emotion and acting on it — the disciplined investor uses the moment of feeling as a signal to slow down, not as a trigger to act. Pre-commitments (the kill-thesis, the quarterly cadence, the position-size hard limits) exist to enforce that gap mechanically when the emotional pull is strongest.