Having trouble? Click to reload

Valuation Score

How do five multiples turn into one number, and what does high mean?

The Valuation Score turns ten readings of price into one number from 0 to 100, where high means cheap against our main comparison pool and against this company's own past. It is a place in a ranking, never a value in dollars.

The formula

Valuation Score = (Sum of 10 percentile readings ÷ 10) × Quality multiplier

  • The ten readings: Five multiples, each read twice. The five are P/E, P/B, P/EBIT, P/R and PEG. Each is read as it stands today, and again as the distance between today and its own 20-quarter average.
  • A percentile reading: Each of the ten becomes a place from 0 to 100 against our main comparison pool, never against the industry. A reading of 90 means cheaper on that one measure than 90% of that pool. A company outside the pool is placed on the same yardstick without joining it.
  • Divided by a fixed 10: Always 10, never however many readings happened to exist. A multiple that cannot be computed counts as nothing, so a company with one usable multiple is pulled towards the bottom rather than flattered by its single reading.
  • The quality multiplier: Taken from the DCA Score, on a straight line from 0.70 at a score of 0 to 1.30 at 100. No DCA Score means 1.0. The answer is then held between 0 and 100.

Example

Two imaginary companies whose ten price readings come out identical, so the market is asking about the same price for each. Company A passes most of the business quality rules. Company B passes few of them.

The price readings said exactly the same thing about both: mildly cheap, 60. They finish 22 points apart because of the business behind the price. That multiplier is the whole reason a cheap weak business does not score as well as a cheap strong one. Note also that the base was divided by 10 in both cases, whether or not all ten readings existed.

What it tells you about the business

It tells you how this price compares with other prices, and with the price this company used to carry. It does not tell you the company is good, and it does not say the price is about to rise.

Why it changes

The price fell, or profits, assets or sales rose while the price stood still. Or its DCA Score improved, which lifts the multiplier.

The price rose faster than the business did. A price that doubles on unchanged profit moves every one of the ten readings the same way.

Five of the ten readings compare the company with its own 20-quarter average, so a company that has been expensive for years slowly stops looking expensive against itself, even with no change in price.

The trap

It says cheap, never worth. There is no discounted cash flow model here, no cost of capital, no fair value and no price target anywhere in MarginX. It compares prices with prices.

Cheap is very often correct. A low price usually reflects something real: a shrinking business, a customer leaving, a debt coming due.

A missing multiple is not a free pass. The divide is always by 10, so one brilliant reading on the only usable multiple still produces a low score. That is the intended answer.

It mixes old figures with a live price. Profit and assets come from the last filing, weeks or months old. The price is today's.

The score is exactly these five multiples. The other price measures on the page, dividend yield, EV/EBITDA and tangible equity against market cap, are there to read beside it and sit outside the arithmetic.

Where you see it in MarginX

A plain verdict sits beside the number on the same cuts, from Very attractive at the top down to Stretched valuation. Open the dialog before you trust either: a middling score from middling prices and a good base cut down by a weak business are completely different situations.

MarginX is a software research tool and data publisher, not an investment adviser. Contact: support@marginx.biz