About Quality Investor
A free, transparent scorecard for public companies built entirely from SEC EDGAR filings. Every number is point-in-time as of the actual filing date — no restated values, no look-ahead bias.
What is this?
Quality Investor scores public companies on profitability (the primary signal) and anti-fragile balance-sheet health (the safety filter), then tracks the directionof each business’s fundamentals over its own multi-year filing history.
The result is a ranked scorecard — currently 464 companies from a 464-name universe — with a Trajectory rating (strengthening, stable, or weakening) that tells you whether the business is getting better or worse on the metrics that matter.
This is a monitoring tool, not a recommendation engine. It surfaces the data; you make the decisions.
Methodology
Primary signal: gross profitability
The ranking uses gross profits divided by total assets (GP/A) — the measure introduced by Robert Novy-Marx in “The Other Side of Value” (Journal of Financial Economics, 2013). Novy-Marx showed that profitable firms earn significantly higher returns than unprofitable ones, and that profitability is complementary to value — not opposed to it.
Secondary filter: anti-fragile safety
Companies pass through a balance-sheet health checklist drawn from the Piotroski F-Score (Piotroski, Journal of Accounting Research, 2000) and supplementary checks on leverage, liquidity, cash-flow quality, and accruals. These filters can only remove names from the ranked list — they never reorder it. The safety filter is a gate, not a signal.
Trajectory rating
The strengthening / stable / weakening badge is computed from ordinary-least-squares slopes over the trailing five fiscal years of seven headline metrics (gross profitability, ROE, gross margin, operating margin, free cash flow, net cash, and share dilution). Each metric votes improving, stable, or deteriorating; the aggregate label requires a supermajority. Red-flag veto gates (unfavorable FCF/net-income divergence, negative accruals) can cap a rating at stable regardless of the vote count.
Trajectory is a monitoring diagnostic of business direction — not a prediction of returns.
Point-in-time discipline
Every fundamental value on this site is anchored to the SEC filing’s filed date — the day the document actually appeared on EDGAR — not the fiscal period it covers and not a later restatement.
This matters because most free financial data providers serve restated or latest values, which embed information that was not available to investors at the time. Backtests and scorecards built on restated data overstate real-world results. The academic term for this error is look-ahead bias, and it is the single largest source of spurious findings in quantitative finance.
McLean and Pontiff (Journal of Finance, 2016) estimated that publication-driven decay erodes roughly 58% of in-sample returns out of sample. Point-in-time discipline does not eliminate this decay, but it removes the largest known source of measurement fraud.
Data source
All data comes from the SEC EDGAR public filing system — specifically the XBRL companyfacts API, which provides structured fundamental data tagged with filing dates.
Scores are refreshed weekly. The pipeline fetches only filings that have changed since the last run (incremental refresh), respecting SEC’s rate guidance.
Financial-sector exclusion:companies with SIC codes 6000–6999 (banks, insurers, REITs) are excluded from the GP/A ranking because gross profit is not a meaningful concept for financial firms. This follows the convention established by Novy-Marx and Fama-French.
What this is not
- Not investment advice. Nothing on this site is a recommendation to buy, sell, or hold any security. The publisher is not a registered investment adviser.
- Not real-time data. Scores reflect the most recent annual or quarterly filing, which may be weeks or months old.
- Not survivorship-free. The universe is drawn from current SEC-registered tickers. Companies that have delisted, merged, or gone bankrupt are not included in the current rankings, which introduces survivorship bias.
- Not a complete picture.Automated XBRL extraction covers roughly 30–40% of the factors an experienced fundamental analyst would evaluate. Qualitative dimensions — competitive moat, management quality, industry dynamics — are not scored.
Coverage
The current universe covers 464 companies. 464 pass the quality and safety gates and appear in the ranked scorecard. Last scored 2026-07-22.