LAB #4919 NEUTRAL: LEDGER BENCH: Apply filter: Session = OFF_SESSION — improves win rate from 31.9% to 41.9% (+10.0pp, — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Session = OFF_SESSION — improves win rate from 31.9% to 41.9% (+10.0pp, n=129) SAME-WINDOW EFFECT: The filter would have blocked nothing in the live window that matters — every listed trade is timestamped 10:25–11:45 CET, which is squarely in-session, so the OFF_SESSION filter is a no-op on this sample. The three CLOSED/LOSS trades (EURUSD, XAUUSD, AUDUSD) and the two CLOSED/WIN trades (USDCHF, USDCAD) would all still have fired, leaving the window's realized result unchanged at 2W/3L. The +10.0pp claim therefore cannot be validated or falsified by the current live trades — it rests entirely on the n=129 backtest subset, which is a different population than what's actually trading. OVER-TIME PROJECTION: If the 41.9% vs 31.9% gap is real and stable, this is a large edge — roughly a 31% relative lift in hit rate, enough to flip a marginal system profitable. But n=129 with a 10pp delta is thin: the sta Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Author: Caldin, Andrew Stewart
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7224994264