Magic Formula investing, explained

Joel Greenblatt introduced the approach in The Little Book That Beats the Market. It ranks companies separately on earnings yield and return on capital, then adds the two ranks. A smaller combined rank means a stronger combination of the two measures.

• Earnings yield = EBIT ÷ enterprise value. EBIT is earnings before interest and tax. Enterprise value reflects the price of the whole business, including debt and allowing for cash. A company with $12 million of operating earnings and a $120 million enterprise value has a 10% earnings yield.
• Original return on capital = EBIT ÷ (net working capital + net fixed assets). This compares operating earnings with the tangible capital used in the business.
• Combined rank = earnings-yield rank + return-on-capital rank. Rank 1 is the highest value on each measure.

Earnings yield here is an operating-profit measure, rather than the inverse of the usual net-income P/E ratio. A strong result on one measure can offset a weaker result on the other.

Assume these ten companies pass the same filters and have complete data. This example uses Stock Meadow's ROIC measure for quality. Each measure is ranked from highest to lowest.

Company A leads: its ranks are 3 and 2, giving a combined rank of 5. With ten peers and no ties on either measure, the yield percentile is (10 − 3) ÷ 9 × 100 = 77.8; the ROIC percentile is (10 − 2) ÷ 9 × 100 = 88.9. Their average is 83.3 out of 100.

Companies B and C tie because each has a combined rank of 6. The score summarises a position among these peers; it is not a forecast of a company's return.

Stock Meadow averages two equally weighted percentile scores: trailing twelve-month operating income divided by enterprise value, and trailing twelve-month return on invested capital (ROIC). Higher values score better on both measures.

The quality measure differs from Greenblatt's original formula. Stock Meadow uses after-tax operating income divided by invested capital (debt plus shareholders' equity). This is an adaptation, so its ranking can differ from the original Magic Formula screener.

1. Choose Magic Formula in the scoring-model picker.
2. Set the company and metric filters you want. Filters run before the scores are calculated.
3. Open a stock's breakdown to see earnings yield, ROIC and the points contributed by each.

The model excludes Financial Services and Utilities, and leaves out stocks missing either scoring measure. At least ten eligible stocks must remain. Metric score switches and custom weights do not change this fixed recipe.

You can compare within each sector. A sector with fewer than five eligible stocks falls back to the whole eligible list. Changing filters or the peer group can change a stock's score even when its financial figures stay the same. Equal values share a midpoint percentile; equal overall scores are ordered by ticker.

See the full methodology for the shared filtering and ranking rules.

• Cheap earnings may not last. A temporary profit peak can make earnings yield look attractive just before business conditions weaken.
• Accounting affects comparisons. Acquisitions, unusual income, taxes and the amount of capital on the balance sheet can alter profitability measures.
• A rank has no valuation target. It does not estimate a fair share price or show whether the whole peer group is expensive.
• Two measures cannot cover every risk. Review debt, cash flow, competitive pressures and the latest reports before drawing a conclusion.

Stock Meadow's scores are research tools based on reported financial data. They do not predict returns or recommend buying a particular stock.

For the original strategy, see Joel Greenblatt's The Little Book That Still Beats the Market and the official Magic Formula Investing FAQ. This guide describes Stock Meadow's implementation; Stock Meadow is independent of those publications and services.

For a different way to examine business quality, read our Piotroski F-Score guide.