ValueScreener

Lesson 1 of 9

What a thesis is

A thesis pairs each reason a business is worth more with the filed figure that would prove it wrong. The price you paid is recorded, never a condition.

Javier Sanz, founder

By Javier Sanz

Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.

Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)

In one minute

  • A thesis pairs each reason a business is worth more with the filed-figure condition that must hold for it to be true.
  • The unit is a pair: a reason with no condition is background, and a condition with no reason is a rule nobody can explain.
  • The price you paid is recorded at the top of the document, as a date, never as a condition.

A thesis is the reasons a business is worth more than the price being asked for it, together with the numeric conditions on filed figures that have to stay true for those reasons to hold. Most written theses contain only the first half, which is an opinion with a date on it and nothing it forbids.

What one pair in a thesis is made of

  1. 1ReasonWhy the business is worth more
  2. 2FigureThe filed number that could prove it wrong
  3. 3LevelThe line that number must stay on the right side of
  4. 4CheckRe-checked every night against the newest accounts
Illustrative round numbers, not any company's filings.

The unit is a pair, not a paragraph

A thesis is a list of pairs, and each pair is one reason together with the condition whose failure would refute it. A reason with no condition is background: it will still be background in three years. A condition with no reason is a screening rule that arrived from somewhere else, and when it breaks there is nothing to go and look at.

Two more fields belong to the document as a whole: the date it was written, and the price on that date.

Why the price is a record, not a condition

A year later there are two different questions: was the reason right, and did the position work. The recorded price is what keeps them apart. Without it, only the second question can be answered, and it teaches little, because everything the market did in between is mixed into the answer.

Graham's line between investment and speculation, in chapter 1 of The Intelligent Investor, is that an investment promises safety of principal and an adequate return, on analysis. The analysis is the part that has to still exist a year later, in the form it was actually held. A thesis is where it is kept.

An unwritten reason gets rewritten

Memory is reconstructive. A decline recruits reasons the reader did not hold at the time, because a losing position demands an explanation. A rise quietly retires the caveats, because a working position does not feel like it needs them. Neither edit registers as an edit, since there is no earlier version to compare against. A dated document is that earlier version.

Klarman describes a related pressure at institutional scale, in Margin of Safety: a manager measured against an index every quarter is pushed toward decisions the same person would not make with their own money over a decade. A private reader feels the same pressure from inside, with nobody watching for it.

A thesis is not a valuation

A valuation

  • A number with assumptions attached: a discount rate, a growth path, a terminal value.
  • An assumption is never refuted. It is replaced by another one, chosen by the same analyst.

A thesis

  • A set of claims with tests attached to filed figures.
  • The company publishes the figure on a schedule it does not control, and the figure either satisfies the condition or does not.

Only one of these is monitored by anything, because only one of them a filing can contradict.

ValueScreener re-checks conditions on filed figures against the newest filed data and sends one message when one moves from holding to broken. A written reason with no figure gets a review date instead.

Check yourself

What are the two parts of a thesis, as this lesson defines it?

  1. A price to sell at and a price to buy more at
  2. A reason the business is worth more, and a condition on a filed figure that would refute it
  3. A forecast of next year's earnings

Show the answer

Answer: B.

A reason with no condition is background. A condition with no reason is an unexplained rule. The pair is the unit.

Why is the price paid recorded at the top of a thesis rather than written as a condition?

  1. Because the price never matters
  2. So that whether the reasons held can be answered separately from whether the position made money
  3. Because prices are not filed anywhere

Show the answer

Answer: B.

Without the recorded price, only "did it work" can be answered, and that answer mixes in everything the market did in between.

What happens to a reason that is never written down?

  1. It stays exactly as it was formed
  2. Memory rewrites it: declines recruit new reasons, rises quietly drop the caveats
  3. It automatically becomes a condition

Show the answer

Answer: B.

Neither edit registers as an edit, because there is no earlier version to compare against. A dated document is that earlier version.

Sources

  • Graham, The Intelligent Investor, revised edition 1973, chapter 1, on investment against speculation.
  • Klarman, Margin of Safety, 1991, on relative performance corrupting the process of a managed portfolio.

What this leaves out

  • This lesson describes the shape of the document, not its contents. It names no reason worth holding and no figure worth conditioning on.
  • Writing a thesis down does not make the reasoning inside it correct. It makes it findable later, in the form it was actually held.
  • The price recorded at the top is a record, not a condition, and no figure in this lesson depends on it.
  • Nothing here is advice about any security.

The method, not the running of it

Lesson 1 of 9 of The investment thesis. It teaches how the step is run. It does not run it on anything you own, and it cannot tell you the day a reason you bought on stops being true.

Free audit
  • What makes a reason falsifiable is the next lesson, and the test that separates a reason you can check from one you cannot.
  • What a discount rate is doing sets out the model a valuation is, and why it is a range rather than a claim anything can refute.
  • About states the failure mode this course was written against, in the founder's own words.

Research tool, not investment advice. See our methodology for how the product reads a filing and what it refuses to compute.

Once a week, what changed in the filings of US-listed companies, why it can matter to a thesis, and what deserves a closer review. Grounded in filings, not market noise.