The F-Score · Journal of Accounting Research, 2000
Nine tests in one score. Its high and low bands can be counted across the US market.
Counted 11 Sep 2026. Numbers are Piotroski's unless marked Ours.
Piotroski classified firms scoring 8 or 9 on the nine tests as high-score firms, and the strategy he tested selected them from inside his high book-to-market group.
8 sits at or above the 95th percentile, 7, the highest published mark.
Companies with filed statements, one each; exact zeros left out as missing data; banks, capital markets firms, equity REITs, insurers and mortgage REITs left out (734); extremes capped at the 1st and 99th percentiles.
Same nine tests, four inputs changed. Return on assets and turnover use year-end assets, not beginning-of-year. The debt test uses long-term debt over year-end assets, not average. Net income replaces income before extraordinary items. No new equity reads as the diluted share count not rising. The cohort is the market less banks, insurers, brokers and real estate trusts.
He classified firms scoring 0 or 1 as low-score firms, and the long and short version of his test held them on the short side.
1 sits on the 5th percentile, and below the 10th, 2.
Companies with filed statements, one each; exact zeros left out as missing data; banks, capital markets firms, equity REITs, insurers and mortgage REITs left out (734); extremes capped at the 1st and 99th percentiles.
Same tests and substitutions as the high group. A company with no prior filed year is still scored, with its six year-on-year tests counted as failed. The cohort also drops every exact zero as missing data, so a company failing all nine tests is not in it, and the share shown counts the companies scoring one.
Each year from 1976 to 1996 he ranked firms by book-to-market, kept the highest fifth, and applied the nine tests only there. The median firm in that group had a book-to-market ratio of 1.721, a price near 0.58 times book. A fifth of the market is a rank, not a level, and he published no book-to-market cut-off.
Joseph Piotroski
The idea is Piotroski's. Every number is ours.
The median firm in his cheapest fifth sat near 0.58 times book, so 1.0 is a loose stand-in. Six of the nine signals compare two years, so a company with one filed year scores zero on those six.
Apple Inc. · AAPL · NASDAQ
Latest annual filing: fiscal 2025. Prices of 10 Sep 2026. Computed 11 Sep 2026.
14-day trial, card at signup.
Research tool, not investment advice. Meeting a test is not a reason to buy.
ValueScreener is not affiliated with or endorsed by Joseph Piotroski.