Cross-sectional momentum: each month, buy what rose most over the last 12 months, skipping the last month. Here in its sector version (Moskowitz and Grinblatt, 1999): the best 3 of 10 US sectors.
Recent winners keep beating losers for 3 to 12 months (long-short spread). The long leg alone promises less.
| Market | Unseen period | Verdict | Rule / buy-and-hold | Max drawdown rule / buy-and-hold | Proof |
|---|---|---|---|---|---|
| Rotation (universe of the record) | 1993-09 → 2026-08 (33.0 y) | No proof of an edge yet | +5,974.9% / +2,894.3% | 39.6% / 42.9% | SAP-635556a921c1 |
What the theory brings, measured. The rule did better than holding the 10 sectors over 33 unseen years, all after the 1993 paper: +5,975% vs +2,894%, with a slightly smaller maximum drawdown (39.6% vs 42.9%), fees included. The gap passes the tests the Lab can compute on a series (SPA p = 0.034; in only 4.7% of draws does the gap vanish). The verdict stays "no proof yet" because two controls (PBO, CPCV) need variants, which a rule fixed in advance does not have: a known limit of the judge, being fixed, not a flaw of the rule. Under the scripts' "fixed rule" reading, the threshold would be much stricter with the trials counted in this wave: so the Lab does not say "survives". What is measured: finance's most replicated prediction held on sectors, 33 years after it was published.
Rule written and preregistered on 12/10/2026, before any proof, with its source's parameters; one trial per market, nothing tuned; judged in certified mode by Lab 270 as a native rotation, which cuts the last months of the series itself; their share was fixed in advance so that the unseen period starts after the source's publication. "Rule / buy-and-hold": gain of the rule and of buy-and-hold on the same bars, 10 bp fee per order. Proofs SAP-635556a921c1. Gaps with the source: Ken French sectors (research baskets, no management fees, not tradable as such) instead of stocks; long only; the list of rules was chosen by an AI that may know their post-publication history (declared in the 12/10/2026 record). Data: Ken French Data Library, FRED (Board of Governors). Research on past data, not advice; no future return is promised.
Jegadeesh, N., Titman, S. (1993), "Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency", Journal of Finance 48(1), 65-91, DOI 10.1111/j.1540-6261.1993.tb04702.x; Moskowitz, T. J., Grinblatt, M. (1999), "Do Industries Explain Momentum?", JF 54(4), 1249-1290, DOI 10.1111/0022-1082.00146.