Piotroski F-Score Explained: 9 Criteria, Calculation, and How to Screen for Quality Stocks

The Piotroski F-Score is one of the most durable quantitative tools in value investing. It reduces a company's financial health to a single number between 0 and 9, derived entirely from public accounting data. No estimates, no forecasts, no analyst opinions. Just nine binary tests applied to the income statement, balance sheet, and cash flow statement.

In the 25 years since Piotroski published his original paper, the F-Score has become a standard component of quality screens used by individual investors, quant funds, and academic researchers. This guide explains every component in plain language, covers the calculation methodology, and shows how to apply the F-Score as part of a disciplined stock research process.

Who Created the Piotroski F-Score?

Joseph Piotroski published 'Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers' in the Journal of Accounting Research in 2000. Piotroski was an accounting professor at the University of Chicago at the time, and the paper grew out of a deceptively simple observation.

Value investors typically screen for low price-to-book (P/B) stocks on the assumption that cheap assets will eventually be repriced by the market. But Piotroski noticed that within a universe of low P/B stocks, financial performance varied enormously. Some companies were cheap because they were genuinely improving their business. Others were cheap because they were deteriorating. Buying the whole low-P/B bucket indiscriminately produced mediocre results because the losers dragged down the winners.

Piotroski's solution was to build a financial strength score that could separate improving companies from declining ones. He tested the system on data from 1976 to 1996 and found that buying high F-Score stocks while avoiding low F-Score stocks within a low P/B universe generated a mean annual return approximately 7.5 percentage points higher than a simple low P/B buy-and-hold approach. The F-Score became a foundational reference in quantitative value investing.

The Three Pillars of the F-Score

Piotroski organized his nine criteria into three groups, each targeting a different dimension of financial health. Every criterion is binary: a company either passes (scores 1) or fails (scores 0). The total F-Score is the sum of all nine binary outcomes.

The three pillars are:

1. Profitability (4 criteria): Is the business generating positive earnings and cash flows?
2. Leverage, Liquidity, and Source of Funds (3 criteria): Is the balance sheet getting stronger or weaker?
3. Operating Efficiency (2 criteria): Is the business becoming more productive?

This structure matters. A company can look profitable on the surface while quietly degrading its balance sheet or issuing dilutive shares. The F-Score catches all three failure modes simultaneously.

All 9 Criteria: Full Reference Table

Pillar 1: Profitability (F1 to F4)

The profitability pillar tests whether a company is generating real economic output today, and whether that output is improving.

F1: Return on Assets (ROA)

ROA is calculated as net income divided by total assets at the beginning of the fiscal year. Piotroski uses beginning-of-year assets rather than average assets to measure how effectively capital deployed at the start of the period generated earnings by year end.

A company earns a 1 for F1 if ROA is positive. Any positive net income passes. The test is intentionally lenient at this step because the broader screen (low P/B stocks) already filters for out-of-favor companies; the question is simply whether they are still profitable at all.

F2: Operating Cash Flow

Net income can be manipulated through accruals, timing of revenue recognition, and non-cash items. Operating cash flow is harder to manufacture. Piotroski scores a 1 if cash flow from operations (CFO) is positive in the current year.

This criterion works alongside F4. Together they test whether reported profits are supported by actual cash generation.

F3: Change in Return on Assets

A positive ROA is good. An improving ROA is better. F3 compares the current year ROA to the prior year ROA. If ROA has increased, the company earns a 1. If ROA has held flat or declined, it scores 0.

A company can pass F1 (positive ROA) but fail F3 (declining ROA), which is a meaningful early warning signal for businesses that peaked and are now drifting.

F4: Accruals (Cash Flow vs. Net Income)

Accruals capture the difference between reported earnings and actual cash generation. When a company reports high profits but low cash flow, it is accumulating accruals: unbilled revenue, extended payables, deferred expenses. Persistent large accruals are often a precursor to earnings restatements or write-downs.

Piotroski operationalizes this as: (CFO / total assets) versus (net income / total assets). If the cash-flow-to-assets ratio exceeds the income-to-assets ratio, the company scores 1. This rewards firms where real cash generation equals or exceeds reported profits.

Pillar 2: Leverage, Liquidity, and Source of Funds (F5 to F7)

The second pillar examines whether the balance sheet is getting healthier and whether the company is financing its operations without diluting shareholders.

F5: Change in Leverage

Leverage is measured as long-term debt divided by average total assets. If that ratio declined year over year, the company scores 1. A falling leverage ratio indicates the company is reducing its dependence on debt, which lowers financial risk and increases operating flexibility.

Rising leverage when fundamentals are already weak signals a company relying on debt to sustain operations it can no longer self-fund.

F6: Change in Liquidity

Current ratio is current assets divided by current liabilities. If the current ratio improved from the prior year to the current year, the company earns a 1 for F6.

A rising current ratio means the company has more short-term assets relative to short-term obligations than it did previously. That improving buffer reduces the probability of near-term liquidity stress and suggests management is building rather than depleting its financial cushion.

F7: Absence of Dilution

If a company issued new common shares during the fiscal year, it scores 0 on F7. If the share count held steady or declined, it scores 1.

Share issuance during a period of financial weakness is a specific warning pattern. Companies under financial stress often issue equity because they cannot access debt markets on reasonable terms, or because they need to shore up their balance sheet. Piotroski treats any new issuance as evidence the company cannot self-fund its operations.

This criterion also protects against equity dilution that destroys per-share value for existing holders.

Pillar 3: Operating Efficiency (F8 and F9)

The final two criteria test whether the business is becoming more efficient at generating revenue and profit from its asset base.

F8: Change in Gross Margin

Gross margin is revenue minus cost of goods sold, divided by revenue. An improving gross margin year over year earns a 1.

Gross margin improvement can come from pricing power, cost reduction, product mix shifts toward higher-margin lines, or operational improvements in production. Whatever the source, an expanding gross margin signals that the core economics of the business are strengthening. A contracting gross margin, even with rising revenue, suggests intensifying competition or cost pressure.

F9: Change in Asset Turnover

Asset turnover is revenue divided by beginning-of-year total assets. If asset turnover improved year over year, the company scores 1.

A rising asset turnover ratio means the company is generating more revenue per dollar of assets deployed. This reflects improved productivity: the business is working its balance sheet harder. Declining asset turnover indicates either that revenue growth is lagging asset growth (often caused by acquisition activity or capital expenditure that has not yet translated into revenue) or that the business is contracting relative to its asset base.

Scoring and Interpretation

Every company receives a total F-Score from 0 to 9, which is the simple sum of all nine binary outcomes.

Piotroski organized the scores into three interpretive bands:

High F-Score: 8 or 9 Companies in this range pass nearly every financial health test simultaneously. Their profitability is real, their balance sheets are improving, and their operations are becoming more efficient. Within a low P/B screen, high F-Score stocks represent the most financially robust candidates.

Middle F-Score: 3 to 7 The majority of companies fall in this range. No simple directional interpretation is reliable here. A score of 5 achieved by passing three profitability tests but failing all balance sheet tests is very different from a 5 achieved by passing evenly across all three pillars. Middle-range scores require reading the individual components, not just the total.

Low F-Score: 0, 1, or 2 Companies scoring 0 to 2 fail nearly every test. Piotroski's original work used these low-score stocks as the short side of a long-short strategy: the 'losers' predicted to underperform because their deteriorating fundamentals would eventually force the market to lower their prices further, even from already-depressed P/B levels. For a long-only investor, a score of 0 to 2 is a significant red flag regardless of how cheap the valuation appears.

Piotroski's Original Backtested Results

In the original 2000 paper, Piotroski tested a binary strategy using data from 1976 to 1996 (20 fiscal years of data tested out-of-sample). The key findings:

The mean return to a strategy of buying high F-Score (8 or 9) low P/B stocks and shorting low F-Score (0 or 1) low P/B stocks was approximately 23 percent per year over the test period. This is not the return to buying high F-Score stocks alone. It is the combined return of the long-short spread.

For long-only investors, the more relevant finding was the improvement in hit rate. Among low P/B stocks, roughly 44 percent of stocks with high F-Scores earned positive returns over the following year, compared to roughly 9 percent of stocks with low F-Scores. The F-Score did not guarantee positive returns, but it substantially shifted the odds.

One important context note: these results come from a specific historical period and a specific market (U.S. equities). Subsequent academic research has confirmed the F-Score's predictive value in a range of international markets, though the return premium has compressed in some markets as the strategy became more widely known.

Combining the F-Score with Value Screens

Piotroski designed the F-Score specifically for use within a low P/B stock universe. The logic is that cheap stocks (low P/B) already have a valuation margin of safety baked in. The F-Score then filters for quality within that cheap universe: you want stocks that are both cheap and improving.

This combination addresses a fundamental problem with pure value investing: value traps. A stock can trade at a low P/B ratio for years while the underlying business deteriorates, and the 'cheap' price simply reflects correctly priced impairment. The F-Score helps identify which cheap stocks are worth holding and which are cheap for good reason.

Common ways investors apply the combined screen:

• P/B below 1.0 + F-Score 7 or higher: A strict filter that targets deeply discounted stocks with strong financial momentum
• P/B below 1.5 + F-Score 6 or higher: A broader filter that captures more candidates while still tilting toward quality
• Bottom quartile P/B by sector + F-Score 7 or higher: A sector-adjusted version that avoids the systematic bias toward capital-heavy industries (financials, utilities) that dominate raw P/B screens

Some investors also combine the F-Score with additional quality metrics: strong free cash flow yield, low price-to-earnings, or improving return on invested capital. The F-Score is additive to these screens because it tests financial trajectory, not just current valuation.

How Modern Stock Screeners Apply the F-Score

Most institutional-grade screeners and quantitative research platforms now include the Piotroski F-Score as a pre-built filter. The implementation is largely standardized because all nine criteria are defined precisely in the original paper and rely entirely on audited financial statement data.

Key things to check when using a screener's F-Score:

Fiscal year timing: The F-Score uses full fiscal year data. Screeners that use trailing twelve month (TTM) data may produce slightly different results than those using the most recently completed fiscal year. Neither approach is wrong, but be consistent across comparisons.

Denominator for ROA and asset turnover: Piotroski uses beginning-of-year total assets as the denominator for F1, F3, F4, and F9. Some implementations use average assets (beginning plus ending, divided by two). This changes the absolute level of each ratio but rarely changes the pass/fail outcome unless the asset base shifted dramatically.

Share count source: F7 (absence of dilution) requires comparing diluted shares outstanding at year end to the prior year. Some screeners use basic shares; some use diluted. Diluted share count is more conservative and better aligned with Piotroski's intent.

Platforms like Equity Rank incorporate the F-Score as part of a multi-factor fundamental model alongside other quality and valuation metrics. Rather than presenting the F-Score in isolation, integrated platforms allow users to weight it alongside proprietary signals to surface research candidates across thousands of stocks simultaneously.

Limitations of the Piotroski F-Score

No single metric, however well-constructed, is complete. The F-Score has several known limitations that serious users should understand.

It is backward-looking. All nine criteria test what happened in the most recently completed fiscal year versus the prior year. The F-Score cannot capture business model changes, new product cycles, management transitions, or macro tailwinds that have not yet shown up in the financial statements.

It was designed for low P/B stocks. Applying the F-Score to growth stocks or high P/B stocks can produce misleading results. A high-growth software company may score low on F7 (because it issues equity to fund growth) and low on F5 (because it takes on debt to scale) while being fundamentally sound. The criteria were calibrated for mature, asset-heavy businesses trading at distressed valuations.

It does not account for industry context. A current ratio that declined from 2.0 to 1.8 fails F6 even if a ratio of 1.8 is perfectly healthy for that industry. An asset turnover increase for a retailer (where high turnover is normal) means something very different than the same increase for a pharmaceutical company (where asset turnover is structurally low).

It is a binary pass/fail system. Each criterion is either 0 or 1 regardless of magnitude. A company whose ROA improved from -0.1 to +0.01 passes F1 and F3. A company whose ROA improved from +8 to +12 percent also passes F1 and F3. The score treats both identically.

Financial statement timing gaps. Quarterly reporters may have significant interim developments (acquisitions, impairments, restructurings) that are not reflected in annual data. Annual-only application can miss important changes that occurred partway through the year.

The F-Score in Context: One Tool Among Many

The Piotroski F-Score earns its place in a rigorous stock research process because it is transparent, replicable, and grounded in public financial statement data. Every calculation is deterministic. There is no subjectivity.

The F-Score is most useful as a filter and a flag. A score of 8 means the company's recent financial trajectory looks healthy across nine standardized tests. A score of 2 means the financials show multiple warning signs that deserve scrutiny before committing capital. Neither outcome is a verdict.

The strongest research processes use the F-Score alongside other valuation and quality metrics: discounted cash flow estimates, earnings quality measures, return on invested capital, and sector-relative comparisons. The F-Score narrows the starting list. Deeper analysis determines which candidates are worth owning.

Understanding how the score is calculated, why each criterion was chosen, and where the system has documented blind spots is what separates investors who use quantitative tools well from those who treat a single number as a conclusion.