LAB #3107 HARMFUL: LEDGER BENCH: Apply filter: GBPUSD LONG @ NY_OPEN with 6 conf — improves win rate from 31.3% to 62.5 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: GBPUSD LONG @ NY_OPEN with 6 conf — improves win rate from 31.3% to 62.5% (+31.2pp, n=8) SAME-WINDOW EFFECT: Over the current live window, this filter would have blocked all GBPUSD SHORT entries (the opposite direction), which account for 3 of the 10 trades shown — all three were CLOSED/WIN. That means the filter would have removed 3 winning trades and left the remaining 7 trades untouched, reducing total wins from 6 to 3 and likely cutting net PnL significantly (the GBPUSD shorts were profitable). The claimed +31.2pp win-rate improvement is based on n=8 backtest samples, which is statistically meaningless — and it directly contradicts the live evidence that GBPUSD SHORTs are currently the engine's strongest performers. OVER-TIME PROJECTION: Over 3-6 months, this filter would systematically exclude a profitable directional bias (GBPUSD SHORT) that has shown repeated wins in the live window. The back Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com