Building an investment thesis

An opinion is "I like this company." A thesis is a specific claim about why an asset is mispriced, what the market currently believes instead, what will force the correction, and what would prove the whole thing wrong. The distance between those two sentences is the work.

Variant perception: the only source of return

Prices already contain the consensus, so returns come only from holding a view that differs from it AND being right. Michael Steinhardt's term for this is variant perception, and it imposes a discipline most ideas fail immediately: state, in one sentence, what you believe that the market does not. "Great company, growing fast" is not a thesis if the price already says so; it is agreement wearing conviction's clothes. The two honest questions behind every thesis: WHY does this mispricing exist (who is the seller and what is their mistake or constraint?), and why am I the one positioned to see it?

Legitimate sources of edge compress to three: informational (you know something others do not, legally: channel checks, filings actually read, domain expertise), analytical (same facts, better interpretation: the market misweights a segment, misses the unit economics), and behavioral/structural (you can act where others cannot: longer horizon, no index constraint, tolerance for boredom or pain).
Common non-edges: headlines, brokerage notes everyone received, the feeling that a stock "has fallen too much", and expertise in the product rather than the business.

The anatomy of a written thesis

The five-line skeleton

1. CLAIM      what is mispriced, and by roughly how much
2. CONSENSUS  what the market believes now (steelmanned, not strawmanned)
3. VARIANT    what you believe instead, and the evidence
4. CATALYST   what forces the repricing, and roughly when
5. FALSIFIER  the observable fact that would prove you wrong

if line 2 is hard to write, you do not know your opponent;
if line 5 is missing, it is faith, not a thesis

Writing it down is not ceremony. A written thesis dated before entry is the only defense against the mind's revisionism: positions drift into being held for reasons invented after the fact, and the document is how you catch the drift. Funds institutionalize this as the investment memo; the format survives at every scale because the failure it prevents is universal.

Catalysts: the thesis's clock

Cheap can stay cheap for a decade; a mispricing needs a mechanism that forces the market to notice. Hard catalysts have dates (earnings, spin-off completions, regulatory decisions, index inclusions, debt maturities); soft ones are processes (estimates revising, a cycle turning, a buyer emerging). A thesis without any catalyst is a bet that the market corrects itself out of fairness, and the market owes you nothing. The catalyst also sets the HORIZON, which sizing and instrument choice inherit: a two-year rerating story expressed through three-month options is a thesis fighting its own clock.

The pre-mortem: attack it before the market does

Steelman the other side. Someone intelligent is selling to you. Write THEIR memo; if you cannot, you have not found the debate, only your side of it.
Imagine failure, then explain it. Assume it is two years on and the position lost 40%. What happened? The three most plausible stories are your real risk list, better than any generic one.
Check the crowding. If the thesis is on every podcast, your variant perception is the consensus, and the exit will be a doorway during a fire.
Locate the disconfirming data. Decide WHERE the evidence against you would first appear (a competitor's pricing, channel inventory, credit spreads) and watch that, not your own confirmation feed.

The falsifier is chosen now, in the calm

The single most protective habit in investing: before entry, write the observable condition that means the thesis is broken, and what you will do when it appears. Chosen in advance it is cheap; improvised mid-drawdown it competes with fear, hope and a sunk cost, and loses. "The thesis is broken when churn exceeds X for two quarters" survives contact; "I'll re-evaluate if things get bad" does not. This line is what separates conviction from stubbornness, which are otherwise identical from the inside.

Grading a thesis before risking money

Can you state the variant perception in one sentence, and would a smart skeptic agree it is actually non-consensus?
Is the claim quantified through valuation work (a football field, a reverse DCF of what the price implies), or is "undervalued" a mood?
Does the catalyst arrive within the horizon your capital can actually hold?
Is the falsifier observable, dated, and written down?
If all four survive: the thesis is ready for sizing, which is the next doc.

Glossary for this guide

Variant perception: A specific belief that differs from consensus, held with evidence, plus a reason the market is wrong. The only source of excess return; the discipline is stating it in one sentence.
Edge: The reason YOU capture a mispricing: informational (legal, real research), analytical (better interpretation), or structural (a horizon or freedom others lack). No edge, no thesis; a headline is not an edge.
Consensus: What the market currently believes, as embedded in the price. The opponent every thesis must name honestly; if you cannot write the other side's memo, you have not found the debate.
Catalyst: The event expected to force the market to reprice toward your view: earnings, a spin-off, a regulatory decision, a contract. Cheap without a catalyst can stay cheap for years; the catalyst is the thesis's clock.
Falsifier: The observable fact, chosen in advance, that would prove the thesis wrong and trigger exit. Deciding it while calm is cheap; deciding it mid-drawdown is expensive, which is why professionals write it down first.
Pre-mortem: Assume the position failed and explain why, before entering. The three most plausible failure stories are the real risk list, and better than any generic one.
Steelmanning: Arguing the opposing case at its strongest, not its weakest. The intellectual habit that separates testing a thesis from decorating it.
Crowding: Too much capital in one trade. Crowded strategies unwind together, briefly correlating at the worst moment; measuring how much of a story is already positioned is half of risk management.
Reverse DCF: Running the machine backwards: instead of estimating value from assumptions, solve for the growth and margins the current price already implies, then judge whether those are beatable. Often more honest than the forward version, because it removes your favorite input.
Time horizon: How long capital can genuinely wait, set by the catalyst and by the holder's real constraints. Instruments and sizing inherit it; a two-year story in three-month options is a thesis fighting its own clock.

Variant perception: the only source of return

Legitimate sources of edge compress to three: informational (you know something others do not, legally: channel checks, filings actually read, domain expertise), analytical (same facts, better interpretation: the market misweights a segment, misses the unit economics), and behavioral/structural (you can act where others cannot: longer horizon, no index constraint, tolerance for boredom or pain).

Common non-edges: headlines, brokerage notes everyone received, the feeling that a stock "has fallen too much", and expertise in the product rather than the business.

The anatomy of a written thesis

The five-line skeleton

1. CLAIM      what is mispriced, and by roughly how much
2. CONSENSUS  what the market believes now (steelmanned, not strawmanned)
3. VARIANT    what you believe instead, and the evidence
4. CATALYST   what forces the repricing, and roughly when
5. FALSIFIER  the observable fact that would prove you wrong

if line 2 is hard to write, you do not know your opponent;
if line 5 is missing, it is faith, not a thesis

Catalysts: the thesis's clock

The pre-mortem: attack it before the market does

Steelman the other side. Someone intelligent is selling to you. Write THEIR memo; if you cannot, you have not found the debate, only your side of it.

Imagine failure, then explain it. Assume it is two years on and the position lost 40%. What happened? The three most plausible stories are your real risk list, better than any generic one.

Check the crowding. If the thesis is on every podcast, your variant perception is the consensus, and the exit will be a doorway during a fire.

Locate the disconfirming data. Decide WHERE the evidence against you would first appear (a competitor's pricing, channel inventory, credit spreads) and watch that, not your own confirmation feed.

The falsifier is chosen now, in the calm

Grading a thesis before risking money

Can you state the variant perception in one sentence, and would a smart skeptic agree it is actually non-consensus?

Is the claim quantified through valuation work (a football field, a reverse DCF of what the price implies), or is "undervalued" a mood?

Does the catalyst arrive within the horizon your capital can actually hold?

Is the falsifier observable, dated, and written down?

If all four survive: the thesis is ready for sizing, which is the next doc.

Glossary for this guide

Variant perception: A specific belief that differs from consensus, held with evidence, plus a reason the market is wrong. The only source of excess return; the discipline is stating it in one sentence.
Edge: The reason YOU capture a mispricing: informational (legal, real research), analytical (better interpretation), or structural (a horizon or freedom others lack). No edge, no thesis; a headline is not an edge.
Consensus: What the market currently believes, as embedded in the price. The opponent every thesis must name honestly; if you cannot write the other side's memo, you have not found the debate.
Catalyst: The event expected to force the market to reprice toward your view: earnings, a spin-off, a regulatory decision, a contract. Cheap without a catalyst can stay cheap for years; the catalyst is the thesis's clock.
Falsifier: The observable fact, chosen in advance, that would prove the thesis wrong and trigger exit. Deciding it while calm is cheap; deciding it mid-drawdown is expensive, which is why professionals write it down first.
Pre-mortem: Assume the position failed and explain why, before entering. The three most plausible failure stories are the real risk list, and better than any generic one.
Steelmanning: Arguing the opposing case at its strongest, not its weakest. The intellectual habit that separates testing a thesis from decorating it.
Crowding: Too much capital in one trade. Crowded strategies unwind together, briefly correlating at the worst moment; measuring how much of a story is already positioned is half of risk management.
Reverse DCF: Running the machine backwards: instead of estimating value from assumptions, solve for the growth and margins the current price already implies, then judge whether those are beatable. Often more honest than the forward version, because it removes your favorite input.
Time horizon: How long capital can genuinely wait, set by the catalyst and by the holder's real constraints. Instruments and sizing inherit it; a two-year story in three-month options is a thesis fighting its own clock.