LAB #1836 HARMFUL: LEDGER BENCH: Apply filter: BTCUSD LONG @ OFF_SESSION with 5 conf — improves win rate from 31.9% 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 31.9% to 100.0% (+68.1pp, n=10) SAME-WINDOW EFFECT: Over the current live window, this filter would have zero impact — none of the 10 listed trades are BTCUSD, and the filter targets only BTCUSD LONG entries during OFF_SESSION with 5 conf. The claimed +68.1pp win-rate improvement is based on n=10, which is statistically meaningless (95% CI roughly ±30pp at best, likely wider given selection bias). If applied live, it would simply block all BTCUSD longs outside the defined session/confidence — likely reducing trade count without touching the PnL of the non-BTCUSD trades shown. Expected PnL delta over this window: $0. OVER-TIME PROJECTION: Over 3–6 months, this filter would likely reduce BTCUSD long frequency dramatically, and the 100% win rate on 10 trades is almost certainly a mirage — small samples with extreme session/confidence combos Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com