LUNARTULIP LAB / AUTHORITY LEDGERHome ↗

FORECAST LEDGER / AGGREGATE OUTCOMES

What happened after
the original calls.

Understand the sample before interpreting the result.

Currently resolved observations come from a reconstructed historical window, testing prior judgments under one consistent rule set. Desk-discipline observations are recorded separately and will be shown independently once enough outcomes have settled.

Resolved / incl. invalidated52
EW directional hit rate47.1%
CSI500 directional hit rate45.7%
CSI500 hit/miss payoff ratio1.42×

UNIVERSE-EW / RESEARCH BENCHMARK

16 hit / 18 miss / 15 flat

Direction-adjusted average excess -0.5%; hit/miss-only payoff ratio 1.01×.

CSI500 / EXTERNAL BENCHMARK

16 hit / 19 miss / 14 flat

Direction-adjusted average excess +1.7%; hit/miss-only payoff ratio 1.42×.

2026-07-2042.9%19 EW observations settled
2026-07-2750.0%33 EW observations settled
2026-08-0345.2%44 EW observations settled
2026-08-1045.2%44 EW observations settled
2026-08-1750.0%30 EW observations settled
How to read these figures

Directional hit rate = hit / (hit + miss). Expired-flat observations are excluded from the denominator; invalidated calls are excluded from the metric set. The threshold is ±5%, with avoid/trim direction inverted. The hit/miss-only payoff ratio excludes flats. The reconstructed window begins on 2026-03-30; current evidence-link completeness is 100%.

The Desk-discipline cohort currently has 1 settled and 25 tracking observations. Reconstructed and discipline cohorts are reported separately; hit rate should be interpreted alongside average excess return, payoff ratio, sample size and time window.