LAB #4025 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 32.5% to 53.7% (+21.2pp, n=268) — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT — improves win rate from 32.5% to 53.7% (+21.2pp, n=268) SAME-WINDOW EFFECT: Applying a USDCHF SHORT-only filter would have excluded none of the 10 live trades above, as none are USDCHF. Therefore, the immediate PnL and win-rate delta over this exact window is zero — no trades are added or removed. The +21.2pp win-rate improvement is purely a backtest artifact on a different instrument, so it offers no corrective value for the current GBPUSD, USDJPY, NZDUSD, AUDUSD, XAUUSD, EURJPY, or EURUSD positions. The only indirect effect would be psychological: you might feel the model is "improved" while live performance remains unchanged. OVER-TIME PROJECTION: Over 3–6 months, this filter would only alter outcomes when the engine generates USDCHF SHORT signals — which are absent from your recent live mix. If USDCHF SHORT signals are rare (n=268 over the backtest window suggests ~9/month), the fi 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: W7221595208
Cites:
Cited by:
Related to: