The screen is grounded in the criteria Benjamin Graham developed over decades: filters designed to identify statistically cheap, financially sound businesses trading at a meaningful discount to intrinsic value.
"Rule No. 1: Don't lose money. Rule No. 2: Never forget Rule No. 1." (Warren Buffett)
Click any card to read the explanation.
Company Size
A minimum market capitalization, so every candidate has the scale, filing history, and financial stability Graham's other tests assume.
GrahamGrade requires a $1 billion market cap minimum. Below that threshold, companies often lack the filing history and balance-sheet stability the rest of the screen depends on. This is a gate, not a scoring factor.
Price-to-Earnings
Low P/E relative to the market and the company's own history. Paying a fair price for earnings power.
If a company earns $1 per share and you pay $10 for the stock, the P/E is 10. GrahamGrade requires a P/E under 12, tighter than Graham's classic defensive 15 ceiling. The lower the P/E, the more earnings you're getting for your money; a high P/E means you're betting on future growth.
Price-to-Book
P/B below thresholds that indicate the market is not pricing in aggressive growth assumptions.
Book value is roughly what a company would be worth if it sold everything and paid all its debts. A P/B below 1.5 means you're paying close to or less than the business's net worth on paper. Graham saw this as a built-in cushion against being wrong.
Current Ratio
Adequate current assets relative to current liabilities. A basic test of near-term financial health.
Divide what the company can convert to cash within a year by what it owes within a year. Graham wanted that ratio above 1.5. A ratio below 1 means the company might struggle to pay its near-term bills, a warning sign regardless of how profitable it looks on paper.
Long-Term Debt vs. Working Capital
Long-term debt held to a cap relative to net current assets, a second, sharper test of financial strength alongside the Current Ratio.
The Current Ratio measures short-term liquidity. This test looks further out: Graham's Enterprising Investor standard caps long-term debt at 110% of net current assets. A company can pass the Current Ratio test and still carry long-term debt that dwarfs its working capital.
Earnings Stability
Consistent positive earnings over a 5-year lookback, with no net decline over that span.
No earnings losses over the past 5 years, with no net decline over that span: flat is fine, an outright decline is not. A company that can't hold positive earnings through a full cycle is harder to value and carries more risk.
Dividend Record
A current dividend payment. Evidence the company returns cash to shareholders, not just reports profit on paper.
You can't fake cash: dividends impose financial discipline. GrahamGrade requires at least one year of payments as a hard gate; no dividend is an automatic fail. Cutting an established dividend is a red flag we weigh in the qualitative review.
Margin of Safety
The composite score reflects the overall discount to estimated intrinsic value, which is the central Graham concept.
Even the best analysis can be wrong. The margin of safety is the gap between what you pay and what the stock is actually worth. Buy at a big enough discount and you can be somewhat wrong and still come out ahead. It's Graham's central idea: price is what you pay, value is what you get.
Simple Businesses
The business economics should be understood from public filings alone. Complexity, opacity, or reliance on management projections are risk factors, not features.
If you can't explain how the company makes money from reading its annual report, that's a red flag. Graham avoided complex financial structures and businesses that required trusting management forecasts. If it's hard to understand, it's hard to value.
The Graham Number
A ceiling price derived from earnings and book value: √(22.5 × EPS × BVPS). The basis for every margin of safety calculation in the report.
Multiply earnings per share by book value per share, then multiply by 22.5, then take the square root. The result is the classic Graham Number, a fair price ceiling from Graham's own Defensive Investor formula.
Most screeners stop at the numbers. Every stock that clears our filter also gets a written qualitative assessment drawn from the actual annual (10-K), quarterly (10-Q), and material event (8-K) filings.
Each biweekly report checks every ticker for new 10-Q, 10-K or 8-K filings since the last report and re-analyzes accordingly, so ratings reflect the most recent filings available at each cycle, not just the last full review.
Questions About the Methodology
Why does GrahamGrade only cover US-listed stocks?
We screen roughly 6,900 stocks in the current US-listed universe, sourced from NASDAQ Trader's listing files and screened using yfinance financial data, updated each cycle. This is a deliberate scope decision, not a data limitation.
US-listed companies file annual reports (10-K), quarterly filings (10-Q), and material event disclosures (8-K) with the SEC through EDGAR. Those documents give us a decade of structured financial data in a consistent format, the same data set Graham's criteria were designed around. We also apply a hard gate: any company with a Variable Interest Entity (VIE) structure automatically fails the screen, regardless of its quantitative results. VIE structures are particularly common among Chinese ADRs and ADSs listed on US exchanges; they represent a legal form where ordinary shareholders hold no direct ownership in the underlying operating business. That risk is not captured by any of Graham's quantitative criteria, so we disqualify them outright.
Expanding to international markets is not ruled out, but the filing standards, data consistency, and VIE screen would need to be reworked for each regime. That is not the current focus.
Ready to see which stocks pass the screen this cycle?
View Plans