Pre-registered 2026-09-28. B/M = be(FYE t-1) / [shares(FYE t-1) x split-adjust x close(June t)]. June formation frozen 12m, ex-financials, winsorized 1/99. Long-only top-10, hold=21, next-open, 10bps, sqrt(252). Split 2016-01-01 (IS 2010-2015 short — noted). Guard PASS. Survivorship: liquid-101 universe.
| Arm | OOS elig | OOS Ret | OOS Sharpe | OOS MaxDD | Turn/yr | CorrMkt | ResidSh | CorrMom | boot-p | Verdict |
|---|---|---|---|---|---|---|---|---|---|---|
| V1 book-to-market | 74 | +287.48% | 0.76 | -43.44% | 0.43 | 0.83 | -0.22 | 0.55 | 0.0074 | REJECT |
| V2 growth(low B/M) | 74 | +661.99% | 0.97 | -40.46% | 0.58 | 0.85 | 0.13 | 0.71 | 0.0006 | PASS |
| V1b b2m (lag-2) | 73 | +351.78% | 0.82 | -43.56% | 0.39 | 0.83 | -0.12 | 0.57 | 0.0042 | REJECT |
IS (2010-2015, short) reference
| Arm | IS Ret | IS Sharpe | IS MaxDD |
|---|---|---|---|
| V1 book-to-market | +139.42% | 0.86 | -16.86% |
| V2 growth(low B/M) | +228.50% | 1.04 | -16.31% |
| V1b b2m (lag-2) | +94.58% | 0.73 | -15.27% |
V1 OOS robustness (DSR, yearly)
- SR 0.76 | PSR 0.99 | DSR(K=6) n/a | haircut 0.54 | sortino 0.95 calmar 0.31 omega 1.16
- yearly Sharpe/ret: 2016:2.21/+39%; 2017:-0.03/-0%; 2018:-1.05/-15%; 2019:2.23/+32%; 2020:0.07/-6%; 2021:1.48/+25%; 2022:0.79/+15%; 2023:0.70/+9%; 2024:0.27/+3%; 2025:1.62/+28%; 2026:3.20/+29%
PASS bar: OOS Sharpe >= 0.50 AND bootstrap p < 0.05 AND residual Sharpe vs market > 0. corr to momentum reported for the diversifier read (value is the classic momentum complement).
Reviewer note
- Two data bugs were caught and fixed mid-study, and they fabricated a fake edge. (1) the
_ADJclosecolumn is split-adjusted, so FYE shares must be scaled by ALL splits after the FYE (s_end/s0), not just to formation — the first run put NVDA/GOOGL in the 'value' book via a 40x market-cap understate. (2) a few tickers (e.g. COP) carry a 1000x unit-corrupted share count. Before the fix value read OOS 1.43 resid +1.03; after, 0.76 resid -0.22. - Value is beta (REJECT). corrMkt 0.83, resid -0.22, positive but weak years (2017/2018/2020 negative). No sleeve.
- V2 growth's 'PASS' is beta-plus, not a new edge. Low-B/M (expensive/growth) long-only OOS 0.97 at corrMkt 0.85 with resid +0.13 is the 2016-2026 growth-era beta — same read as expand7/9 (fundamentals premia are equity beta on liquid large-caps). No promotion.