Magic Formula: Good and Cheap
Rank companies by quality and price, then buy the strongest combination.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 99%
The Magic Formula operationalizes the idea that an attractive investment should be both a good business and a cheap security. Cheapness is approximated by earnings before interest and taxes divided by enterprise value, which asks how much operating earning power the buyer receives relative to the all-in purchase price. Quality is approximated by return on tangible capital, which measures how efficiently working capital and fixed assets become earnings. Every eligible company receives a separate rank for each measure, and those ranks are combined. The strongest combined scores identify businesses offering the best balance of earning power, capital efficiency, and price. This is a screening and portfolio discipline rather than a claim that two ratios capture every business nuance. Its power comes from applying economically meaningful criteria consistently across a broad universe instead of following stories or recent price action.
Origin
Greenblatt and Rob Goldstein tested crude measures derived from Graham's cheapness and Buffett's business quality, then published the unexpectedly strong result in The Little Book that Beats the Market.
Core principles
- 01Good businesses convert tangible operating capital into earnings efficiently.
- 02Cheap businesses offer high earnings relative to the total purchase price.
- 03Combining quality and value is stronger than relying on either factor alone.
- 04Ranking a broad universe imposes discipline and comparability.
- 05Simple metrics can demonstrate robust economic concepts without pretending to be perfect valuations.
How to run it
- 1
Define the universe
Choose a consistent investable universe and apply any necessary size, liquidity, or sector exclusions before calculating rankings.
Pro tip Keep the universe stable so changing inputs do not silently alter the strategy.
Watch out Tiny or illiquid companies may make historical results difficult to reproduce.
- 2
Rank cheapness
Calculate earnings before interest and taxes relative to enterprise value, then rank companies from highest to lowest yield.
Pro tip Use enterprise value so debt and cash are reflected in the all-in acquisition price.
Watch out Unusual earnings or inconsistent accounting can distort the measure.
- 3
Rank business quality
Measure how effectively each company converts tangible operating capital into earnings and rank the universe from highest to lowest return.
Pro tip Normalize working capital and fixed assets when one-off events make them unrepresentative.
Watch out A temporarily high return may not be durable.
- 4
Combine the ranks
Add or otherwise combine each company's cheapness and quality positions. Sort the resulting scores to identify the best joint candidates.
Pro tip Preserve the separate rankings rather than letting one extreme raw ratio dominate.
Watch out Do not optimize weights repeatedly against historical data.
- 5
Build a diversified portfolio
Select a group of the highest-ranked eligible companies rather than betting on one result. Apply the same position-sizing rule across selections.
Pro tip Stagger purchases if that makes the process easier to maintain consistently.
Watch out Concentration can turn a robust average signal into company-specific risk.
- 6
Rebalance patiently
Review holdings on a fixed schedule and refresh the portfolio from the latest rankings. Allow enough time for valuation differences to matter.
Pro tip Document the schedule in advance to reduce emotional overrides.
Watch out Abandoning the method after a short weak period defeats its long-horizon premise.
In the wild
Retailer A earns £20 on £40 of tangible operating capital and is available at an enterprise value of £250. Retailer B earns only £5 on the same capital but trades at £100. Ranking the full universe reveals whether A's superior capital efficiency compensates for its higher price and whether either company offers a strong combined score rather than relying on one attractive ratio.
→ The investor compares price and business quality systematically before selecting candidates.
Common mistakes
Buying on cheapness alone
A statistically cheap company can remain a poor investment when its business consumes capital at low returns.
Paying any price for quality
A strong business can still produce a weak investment return if the purchase price embeds excessive expectations.
Overfitting the formula
Repeatedly changing definitions until a backtest looks ideal replaces the framework's economic logic with data mining.
Is it for you?
Best for
It is best for investors seeking a systematic starting point for value-oriented stock selection.
Not ideal for
It is not ideal for investors who cannot tolerate periods of underperformance or who treat crude accounting metrics as complete valuations.
From the transcript
“the magic formula only really has two parts. Cheap.”
“I ranked all companies based on how cheap they were, and then I took another ranking of how good they were, what kind of returns…”
“all things being equal, if I can get the same price, I'd much prefer to own the business that can reinvest its money in fifty…”
From the episode
Joel Greenblatt Discusses the Thrill of Investing
Joel Greenblatt