FoxAlgo → Research → Academic Audit → Piotroski F-Score

Academic audit

Failedequity XS fundamental

Piotroski F-Score

The three-gate gauntlet · genuine only if it clears all three and survives adversarial refutation
Gate 1
Survivorship-free
free
clean panel
Eliminated here
Gate 2
Placebo ≥ P95
P0
outranked ~0 of 200 baskets
Gate 3
Cost-aware net
RF -0.43
net-negative after costs
Failed
Rejected at the luck gate — its net ranked no better than random baskets (below the P95 skill line).

The Piotroski F-score, from Piotroski (2000) in "Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers", scores each firm 0-9 on nine binary tests of financial health (profitability, leverage/liquidity, operating efficiency) and buys the high scorers, on the finding that within cheap value stocks the score separates future winners from losers.

What we found

Two problems, one on each construction. Tested as a market-neutral long/short spread on the broad survivorship-free panel it is net-negative after costs (RF -0.43). And the canonical LONG-ONLY version — the way the score is usually deployed — turns out to be pure equity beta: it lands at the 0th percentile of random long-only baskets (placebo P0), meaning random selection did better than the F-score. So on the broad panel the score adds no selection skill; its returns are the market's, not the screen's. This is a failed screen, not a diversifying factor-leg.

How we tested it
2005–2026 test windowmodelled liquidity-aware costssurvivorship free
  • Data: survivorship-free 1077-name US common-stock panel, 2005-2026. Liquidity-aware modelled costs.
  • Placebo / robustness test: real result vs 200 random baskets (real vs the random-basket percentile). The long-only F-score basket placed at the 0th percentile.
Source: Piotroski (2000), "Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers", Journal of Accounting Research
Find the paper (Google Scholar) ↗
← The Academic Audit — all 57 studies

Research, not investment advice. “Validated” factor-legs are market-neutral diversifying building blocks with a losing worst year — none is a standalone tradeable strategy. Metrics are cost-aware and modelled (not live fills); the 2005–2026 test window is out-of-sample versus the source paper. Dollar figures are not returns and are omitted by design.