LAB #4165 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = DISTRIBUTION — improves win rate from 32.6% to 37.6% ( — E8 Intelligence Research

@misc{caldin20264165,
  title = {{LAB #4165 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = DISTRIBUTION — improves win rate from 32.6% to 37.6% ( — E8 Intelligence Research}},
  author = {Andrew Stewart Caldin},
  howpublished = {Zenodo},
  year = {2026},
  month = oct,
  doi = {10.5281/zenodo.23446616},
  url = {https://zenodo.org/records/23446616},
  note = {CC BY 4.0},
}

TY  - UNPB
TI  - LAB #4165 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = DISTRIBUTION — improves win rate from 32.6% to 37.6% ( — E8 Intelligence Research
AU  - Andrew Stewart Caldin
PY  - 2026
DA  - 2026/10/18
PB  - Zenodo
DO  - 10.5281/zenodo.23446616
UR  - https://zenodo.org/records/23446616
AB  - IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Quarterly phase = DISTRIBUTION — improves win rate from 32.6% to 37.6% (+5.0pp, n=109)  SAME-WINDOW EFFECT: Applying the DISTRIBUTION-phase filter to the current 10-trade window would have blocked all trades opened on 2026-09-01 and 2026-09-02 (6 trades), since those dates fall in a likely DISTRIBUTION phase per your quarterly cycle. Of those blocked, 2 were wins (GBPUSD SHORT, NZDUSD SHORT) and 4 were losses/opens — netting a small positive effect (+2 wins removed, −4 losses removed, net PnL roughly +2R). The remaining 4 trades from late August (all rizzy5m/7m) would still stand, with 1 win and 3 losses, so the window win rate would rise from ~33% to ~25% (1/4) — actually worse in raw rate, but PnL improves because you cut the bigger losing cluster. Net expected delta: roughly +1.5R to +2.5R over this specific window, driven by avoiding the wkjudas cluster, not by the filter's general logic.  OVER-TI  Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
ER  -