Pre-registered 2026-09-09. Composite = mean z of GPOA/ROE/ROA/CFOA/GMAR/ACC (GPOA + >=4/6 required), ex-financials, winsorized inputs, June formation frozen 12m. Long-only top-10, hold=21, next-open, 10bps, sqrt(252). Split 2016-01-01. Guard PASS.
| Arm | OOS elig | OOS Ret | OOS Sharpe | OOS MaxDD | Turn/yr | CorrMkt | ResidSh | CorrMom | boot-p | Verdict |
|---|---|---|---|---|---|---|---|---|---|---|
| QMJ-profit | 66 | +411.19% | 0.90 | -29.84% | 0.93 | 0.90 | -0.12 | 0.71 | 0.0022 | REJECT |
IS (2010-2015, short) reference
| Arm | IS Ret | IS Sharpe | IS MaxDD |
|---|---|---|---|
| QMJ-profit | +172.08% | 1.02 | -13.45% |
Bar (pre-registered): OOS Sharpe >= 0.50 AND bootstrap p < 0.05 AND resid Sharpe vs market > 0. Full QMJ growth legs deferred (IS ~1y).
Verdict (2026-09-09) — REJECT (beta; quality composite closes the program)
QMJ-profit 0.90 (p=0.0022, IS 1.02 consistent) with resid -0.12, corrMkt 0.90, corrMom 0.71: six metrics jointly still cannot manufacture beta-independent returns long-only on liquid large-caps. Four fundamental arms now read identically — AG/Q2/QMJ-profit beta-reject, accruals marginal-beta-plus, tilt reject. The quality premium, to the extent it exists here, IS equity beta. No sleeve, no tilt, no live change. Fundamentals program closed except pre-registered-new-hypothesis work (e.g. issuance/buyback legs, which need additional tags + their own protocol).