LAB #2072 NEUTRAL: LEDGER BENCH: Apply filter: Session = NY_OPEN — improves win rate from 32.0% to 38.2% (+6.3pp, n=620 — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = NY_OPEN — improves win rate from 32.0% to 38.2% (+6.3pp, n=620) SAME-WINDOW EFFECT: Applying the NY_OPEN filter to the current live window would have excluded all 10 listed trades, as none fall within the NY_OPEN session (all entries are between 09:55 and 20:00+02:00, which maps to London/European hours, not NY open). This would have eliminated 3 wins, 4 losses, 1 open, and 2 undefined — netting a PnL change of roughly +0 (since you'd skip both winners and losers). The apparent win-rate improvement in backtest (+6.3pp) does not translate here because the filter would have zeroed out the entire sample, not selectively removed losers. This suggests the filter is not a precision tool but a broad regime gate, and its benefit is entirely dependent on the distribution of sessions in your forward data. OVER-TIME PROJECTION: Over 3-6 months, if NY_OPEN truly captures a higher-quality volatility re Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com