LAB #4614 PROMISING: LEDGER BENCH: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 32.2% to 44.5% ( — E8 Intelligence Research

@misc{caldin20264614,
  title = {{LAB #4614 PROMISING: LEDGER BENCH: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 32.2% to 44.5% ( — E8 Intelligence Research}},
  author = {Andrew Stewart Caldin},
  howpublished = {Zenodo},
  year = {2026},
  month = oct,
  doi = {10.5281/zenodo.23587342},
  url = {https://zenodo.org/records/23587342},
  note = {CC BY 4.0},
}

TY  - UNPB
TI  - LAB #4614 PROMISING: LEDGER BENCH: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 32.2% to 44.5% ( — E8 Intelligence Research
AU  - Andrew Stewart Caldin
PY  - 2026
DA  - 2026/10/23
PB  - Zenodo
DO  - 10.5281/zenodo.23587342
UR  - https://zenodo.org/records/23587342
AB  - IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 32.2% to 44.5% (+12.3pp, n=137)  SAME-WINDOW EFFECT: Applying the ACCUMULATION-phase filter to the current 10-trade window would have rejected the 3 open/rejected trades (USDCHF LONG, EURJPY SHORT, AUDUSD LONG) and likely the 3 closed losses (GBPUSD LONG, USDJPY SHORT, AUDUSD SHORT) if they fell outside that phase — but since we lack phase labels per trade, the filter would have removed at least the 3 rejected entries, which are currently unrealized. Among the 7 closed trades, 4 were wins and 3 were losses; if the filter had removed 2 of those 3 losses (plausible given ACCUMULATION bias), win rate would rise from 57% (4/7) to ~80% (4/5) on closed trades. However, it may also have removed 1–2 winners, so net PnL delta is uncertain but likely positive given the backtest's +12.3pp improvement. The key risk is that the 3 rejected trades are still ope  Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
ER  -