ValueScreener

Lesson 5 of 6

A margin against your own number

Whose value the margin of safety is measured against here, what happens when that value changes, and the two things it is never measured against.

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

  • The margin of safety here is the discount of a price to the investor's own value: a target price they set, or a model they saved.
  • Measured against a range rather than a point, it becomes a question about whether the gap survives every cell of the grid.
  • It is never measured against a vendor estimate, an analyst consensus, or the price you paid.

The model has produced a range. A margin of safety is what turns a range into a question a reader can answer, and the whole of its meaning depends on whose number is in the denominator.

The margin against the grid rather than against a point

Answerable

  • Is the price below every cell in the grid.
  • Is the price above every cell in the grid.
  • By how much, at the least favourable cell.

Not answerable

The grid does not settle it

  • Whether the price is right.
  • Which cell is correct.
  • Whether the gap will close.

The left column is a question the grid can settle. The right column is one it cannot.

Illustrative round numbers, not any company's filings.

The arithmetic

The margin of safety is the discount of the price to the value: value less price, over value.

A value of 100 against a price of 75 gives a margin of 25 percent. The same price against a value of 120 gives 37.5 percent, and the price did not move. The denominator is doing the work, which is why whose number it is matters more than the arithmetic does.

Whose number

In ValueScreener, exactly one of two things and nothing else: a target price the investor sets on the thesis, or the most recent discounted cash flow they saved.

Three things it is never measured against. A vendor's own value estimate and the margin computed from it, both of which were retired on 2026-09-11 because nobody could replay them and one of them was blind to share splits. An analyst consensus figure. And the price the investor paid, which is a profit and loss figure and says nothing about value.

Against a range, not a point

A model that produced nine cells should not be collapsed to one before the margin is computed. The useful form of the question is whether the gap survives the least favourable cell.

Against the grid of the previous lesson, whose cells span 96.56 to 136.60 dollars a share, a price of 60 gives a margin of 37.9 percent at the least favourable cell and 56.1 percent at the most favourable. A price of 115 gives a negative margin at four cells and a positive one at five, which is not a margin at all but a coin flip between assumptions.

The honest weakness

A margin of safety is only as good as the value behind it, and the value is mostly three assumptions. A reader who wants a comfortable margin can produce one by lowering the required return, and the model will not object.

The defence is the grid and the sourcing discipline of the second lesson: every input listed by kind, the three choices separated from the two readings, and the range reported rather than the point. None of that makes the value right. It makes it checkable, which is the most any of this can offer.

Check yourself

What is the margin of safety measured against here?

  1. The price the investor paid
  2. The investor's own value: a target price they set, or a model they saved
  3. A published consensus estimate

Show the answer

Answer: B.

The price paid is a profit and loss figure. The vendor estimate and the margin computed from it were both retired on 2026-09-11.

A price gives a negative margin against four cells of a grid and a positive one against five. What is that?

  1. A thin margin of safety
  2. Not a margin at all, but a coin flip between assumptions
  3. A signal to lower the required return

Show the answer

Answer: B.

The useful form of the question is whether the gap survives the least favourable cell. Here it does not survive four of nine.

An investor revises their own model upward and the margin rises. What is sent?

  1. A breach email in the other direction
  2. Nothing; the rail is re-armed silently from the new reading
  3. An inbox note recording the revision

Show the answer

Answer: B.

The movement was the investor changing their own assumptions, which is not news to them. Only a price moving against a committed value is what the rail is for.

Sources

  • The product's own methodology page, for what the margin of safety is measured against and for the states a margin rail can be in.
  • The grid in the previous lesson, computed from Toll Brothers, Inc., annual reports on Form 10-K for fiscal years 2017 through 2025.

What this leaves out

  • The prices in this lesson are illustrative figures chosen so the arithmetic can be checked by hand. No company's price is described by any of them and none is a level anybody should adopt.
  • A margin of safety is only as good as the value behind it. A model with an optimistic required return produces a comfortable margin and nothing in the arithmetic objects.
  • Nothing here is a view on any security, and none of it is advice.

A step of the work, on one company

Lesson 5 of 6 of From a filing to a value. It is one step of the method, done on filings that were current when it was written. The step transfers to your own company; the figures in it do not.

Free audit

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.