Systematic & Quantitative Valuation July 5, 2026

The Little Book That Still Beats the Market

Joel Greenblatt · first published 2010

Greenblatt compresses value investing into two numbers: how cheap a business is (earnings yield) and how good it is (return on capital). Rank on both, buy the best combined ranks, repeat mechanically.

The big picture

The book's core bet is that a two-factor ranking — buy good businesses at cheap prices — captures most of what security analysis delivers, without the analysis. Cheapness is earnings yield (operating profit against what the whole enterprise costs); quality is return on capital (operating profit against the capital actually tied up). Rank every stock on both, add the ranks, buy a basket of the best, rebalance yearly.

Why it matters now: in an AI market where quality and price have drifted far apart, a mechanical rank on both dimensions is a fast way to see which names you pay up for and which combine both virtues — before any narrative enters the room.

Two ranks, one sum: quality alone overpays, cheapness alone buys junk — the formula only transacts where both columns agree.

The 3 strategic pillars

Earnings yield, not P/E

Cheapness measured as EBIT against enterprise value — debt included, so capital structure cannot flatter the number.

EY = EBIT / EV. Comparable across leverage levels, which is exactly where plain P/E breaks down.
Return on capital, not ROE

Quality measured as EBIT against tangible capital employed — the profit the operating machine earns on what it truly needs.

ROC = EBIT / (net working capital + net fixed assets). Strips out goodwill and financing effects that distort ROE.
The combined rank + discipline

Neither factor alone — the sum of both ranks, bought as a basket and held through the painful years.

20–30 names, annual rebalance. The formula's edge survives because it underperforms often enough that most people abandon it.

What a Closelook reader does with it

The working use is a ranking pass over any universe you follow: compute both ratios, rank, add. The mistake it prevents is one-dimensional buying — cheap junk (high yield, terrible returns on capital) or great businesses at any price (superb ROC, no yield). The combined rank forces every candidate to clear both bars at once, and the basket-plus-rebalance discipline removes the exit decision where most of the damage usually happens.

The bridge to the Closelooknet approach

Closelooknet's Company Scoring System is in effect a many-factor cousin of this book: nightly cross-sectional ranks over the 334-name universe, with quality and valuation as separate modules that never get blended into a single opaque grade. Greenblatt is the minimal version of the same idea — two columns instead of seven modules. The Valuation Gap framework covers the cheapness half; the asset pages carry the fundamentals both ratios need directly under the chart. Run his two-column rank over an index roster and you have a second opinion on our scores from a 20-year-old formula.

Action-Kit — from theory to practice

Tooling & data

What you need	Where to get it	Cost
EBIT, enterprise value, working capital, net PP&E per company The four inputs both ratios need	stockanalysis.com (statements tab) or company filings EV = market cap + total debt − cash; all four figures sit on the face of the statements.	Free
Screener with EV/EBIT and ROC filters Run the ranking over a whole market instead of a hand list	Finviz (US) or the TradingView screener (global) Free tiers approximate with EV/EBITDA and ROIC; the pack computes the book-exact versions from raw inputs.	Freemium

The formulas

Earnings yield
EY = EBIT / EV, EV = market cap + total debt − cash
- EBIT — trailing operating income
- EV — enterprise value
Higher is cheaper. Debt in the denominator keeps leveraged names honest.
Return on capital
ROC = EBIT / (NWC + net fixed assets)
- NWC — current assets − current liabilities (ex cash/debt in the strict version)
- Net fixed assets — PP&E after depreciation
Excludes goodwill on purpose: the question is what the machine earns, not what someone once paid for it.
Magic Formula rank
Rank = rank(EY) + rank(ROC), ascending — lowest combined rank wins
- Both ratios across the same universe
Exclude financials and utilities (the ratios lose meaning) and set a market-cap floor.

Applied Pack · free members

Greenblatt Applied Pack

The two-column rank as a working sheet and a screener script — feed it a universe, get the combined-rank buy list the book describes.

Greenblatt_Ranker.xlsx — enter EBIT, EV components and capital per name; the sheet computes both ratios, both ranks and the combined Magic-Formula rank with live formulas
magic_formula.py — stdlib-only screener: fundamentals CSV in, ranked buy list out, with the financials/utilities exclusion and market-cap floor as flags
fundamentals_sample.csv — example input showing the expected columns
README.txt — column reference, sourcing tips and the educational-use disclaimer

Pack security

Macro-free Excel · plain-text Python you can read before you run it · no installers, no network access — the code works only on files you provide. Served only from closelook.net; we never distribute through download portals or email attachments. How to verify in 30 seconds →

SHA-256 d18f52cbcb0b4fd4642fd23b65a92fdd9c1e83966900b534524f70f7559e746d

The pack is a free-membership download — no card, free forever. Membership also delivers the Daily Pulse and Weekly Signal to your inbox.

Get free access → Already a member? Sign in →

Closelook publishes a market diary, not investment advice. This condensed read restates the book's ideas in our own words for education — for the author's full argument, go to the source.