← All recordsSW-0030 · Crypto · 2019-02-01 to 2025-06-30

Cross-sectional reversal

Void
90Versions testedevery combination, fixed in advance
+0.00Best versionreturn adjusted for risk, per year
+0.00Luck alonewhat the best scores if nothing works
0.500Scorehas to reach 0.950

Result against chance

Best version found+0.00What luck alone gives+0.00Smallest edge visible+0.98
If the best version cannot beat what luck alone produces from the same number of tries, there is nothing there. The third bar is the smallest real edge this test could have detected: anything below it is invisible to the test, and that limit is stated rather than hidden.

Spread across every version

+0.00median +0.00+0.00zero
Every version tested, worst to best. The box holds the middle half. Showing only the best bar is how noise gets sold as skill. 0% of versions finished above zero.

Why this record is void

run_xs_sweep.py was OOM-killed partway through; finalize.py then scored an all-zero returns matrix, which produces DSR exactly 0.5 and prints NO EDGE FOUND. That output is indistinguishable from a genuine null result and is not one. A guard now refuses to score a matrix where every variant has Sharpe exactly zero.

Notes

The failure mode mattered more than the sweep: a crashed run that looks like a finding is the worst thing this registry could publish.

Sealed on the Bitcoin blockchain before the test ran.
sha256 331fc9fd3d19e76d172da11d4aeb794637fe348efb3a9111a9f06cad238bc088
← SW-0029SW-0031 →