F-0005 · first recorded 2026-08-20
Daily-horizon directional tests on one instrument cannot detect realistic edges; the region is unpowerable, not empty
MEASURED
| Observation | detecting a 52% hit rate needs 4,900 independent days (13.4 years); 51% needs 19,600 (53.7 years) |
|---|---|
| Sample | standard power arithmetic applied to the project's own recorded sample sizes |
| Method | 80% power, alpha .05 two-sided, on a proportion, with observations counted as INDEPENDENT rather than raw |
| What is not established | arithmetic, not an estimate; the only judgement is what effect size counts as realistic |
| Who else publishes this | n/a |
| Evidence | research/POWER-AUDIT-the-daily-lane-is-unpowerable.md |
GEN-1 could resolve nothing below a 54.67% hit rate and measured 52.10% — the best result in the experiment sat below the threshold the experiment could detect. ECON-1’s declared read is 90 independent days, which resolves nothing below 58.5%, so it returns a null in November whether or not the effect is real, and K2 closes the directional programme on that null.
What it does not mean. It does not mean there is no edge. It means the project reported “no edge” where the truthful statement was “no edge larger than about five percentage points.” Conclusions resting on measured magnitudes are unaffected — market making died on 0.00154 bps against 5.19 bps of cost, which is a ratio, not a p-value.