The "Sell in May and go away" saying as a rule: invested from early November to end of April, out from May to October.
Winter returns exceed summer returns in most countries. The paper does not promise to beat buy-and-hold after fees.
| Market | Unseen period | Verdict | Rule / buy-and-hold | Max drawdown rule / buy-and-hold | Proof |
|---|---|---|---|---|---|
| Asia-Pacific ex Japan | 2013-09 → 2026-08 (13.0 y) | No proof of an edge yet | +87.4% / +107.8% | 37.1% / 37.2% | SAP-3c3916056432 |
| Europe | 2013-09 → 2026-08 (13.0 y) | No proof of an edge yet | +143.6% / +176.1% | 35.7% / 35.7% | SAP-f0939b09d16c |
| Japan | 2013-09 → 2026-08 (13.0 y) | No proof of an edge (Deflated Sharpe below 50%) | +62.7% / +206.9% | 26.9% / 32.9% | SAP-a4201f59d0f3 |
| North America | 2013-09 → 2026-08 (13.0 y) | No proof of an edge yet | +101.6% / +432.0% | 34.9% / 34.9% | SAP-1d62b118f584 |
What the theory brings, measured. Invested half the time (50% of days), the "winter only" rule earned +143.6% in Europe vs +176.1% for the whole year, over 13.0 unseen years. It stays below buy-and-hold in all four regions, with the widest gap in North America (+101.6% vs +432.0%). What the saying brings, measured: a large share of the rise with half the time in the market; no extra gain. Only about 13 cycles: the Lab cannot yet tell a real effect from luck.
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, which cuts the unseen period itself: everything 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-3c3916056432, SAP-f0939b09d16c, SAP-a4201f59d0f3, SAP-1d62b118f584. Gaps with the source: Ken French regional indices in dollars instead of national indices; cash at 0% in summer; about 13 cycles in the period, too few to tell a real effect from luck. Data: Ken French Data Library, FRED (Board of Governors). Research on past data, not advice; no future return is promised.
Bouman, S., Jacobsen, B. (2002), "The Halloween Indicator, “Sell in May and Go Away”: Another Puzzle", American Economic Review 92(5), 1618-1635, DOI 10.1257/000282802762024683; Andrade, Chhaochharia, Fuerst (2013), FAJ 69(4), 94-105, DOI 10.2469/faj.v69.n4.4.