LAB #2654 NEUTRAL: LEDGER BENCH: Apply filter: BTCUSD LONG @ OFF_SESSION with 5 conf — improves win rate from 31.1% to — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: BTCUSD LONG @ OFF_SESSION with 5 conf — improves win rate from 31.1% to 100.0% (+68.9pp, n=10) SAME-WINDOW EFFECT: Over the current 10-trade window, this filter would have excluded the 4 BTCUSD LONG trades that occurred during OFF_SESSION with 5 conf — none of which appear in your live list (your live trades are FX/metals, not BTC). So zero trades would have been altered; the 31.1%→100% win-rate claim is based on a different, non-overlapping sample (n=10 BTCUSD only). The live window shows 6 wins/4 losses (60%), and this filter would not have changed that outcome. The +68.9pp improvement is an artifact of a tiny, isolated subset, not a systemic edge. OVER-TIME PROJECTION: Over 3–6 months, applying this filter would likely reduce BTCUSD trade frequency dramatically (OFF_SESSION + 5 conf is a narrow slice). The 100% win rate on n=10 is statistically meaningless (p≈0.001 by chance if base rate is 31%, Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com