LAB #969 NEUTRAL: LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.6% to 33.2% (+1.6pp — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.6% to 33.2% (+1.6pp, n=1328) **SAME-WINDOW EFFECT:** Applying the HIGH pool density filter (≥90) to the live window would have rejected the EURUSD SHORT (rizzy, loss) and the GBPUSD LONG (wkjudas, loss), as both likely originated from lower-density pools. It would also have rejected the XAUUSD SHORT (rizzy5m, still open/rejected) and the EURJPY SHORT (fade_h4, open), but those outcomes are uncertain. The two removed losses improve the observed win rate from ~60% (6 wins out of 10 total trades, ignoring open) to ~75% (6 wins out of 8 closed trades), but the small sample (n=10) makes this noisy. PnL delta is positive but trivial — roughly +2R from avoiding two losses. **OVER-TIME PROJECTION:** Over 3–6 months, the +1.6pp win-rate improvement from the backtest (n=1328) suggests a modest but stable edge. The filter likely reduces false signals fro Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com