LAB #4429 NEUTRAL: LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.5% to 33.0% (+1.4pp — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.5% to 33.0% (+1.4pp, n=1907) SAME-WINDOW EFFECT: Applying the pool-density filter (≥90) would have excluded the three wkjudas entries (GBPUSD LONG, USDJPY SHORT, NZDUSD SHORT) — all three closed as LOSSES. That removes 3 losses from the 10-trade window, leaving 4 wins, 1 loss, and 2 rejected/open trades. Win rate on closed trades would jump from 4/8 (50%) to 4/5 (80%), and PnL would improve by roughly +3R (assuming each loss ≈ −1R). However, the two rizzy7m trades (AUDUSD SHORT loss, USDCHF LONG win) would remain, so the filter's benefit here is concentrated in excluding wkjudas signals — not a uniform density effect. OVER-TIME PROJECTION: Over 3–6 months, the +1.4pp win-rate gain (n=1907) is statistically thin — the standard error on that delta is ~1.1pp, so the true effect could be near zero. The filter likely reduces trade count by ~15–20%, whi Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com