LAB #2129 NEUTRAL: LEDGER BENCH: Apply filter: SOLUSD LONG — improves win rate from 32.0% to 34.0% (+2.0pp, n=50) — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: SOLUSD LONG — improves win rate from 32.0% to 34.0% (+2.0pp, n=50) SAME-WINDOW EFFECT: Applying a SOLUSD LONG-only filter would have zero effect on the current live window — none of the 10 listed trades are SOLUSD, and the filter does not alter entry logic for other instruments. The claimed +2.0pp win-rate improvement (32%→34%) is based on a backtest sample of n=50, which is statistically fragile (standard error ≈ 6.6pp), meaning the observed delta is well within noise. Expected PnL delta over this window: $0.00, with no trades added or removed. OVER-TIME PROJECTION: Over 3–6 months, if the filter is genuinely selective for SOLUSD long setups, it would only affect a small subset of trades (SOLUSD is not in your live list above), so aggregate portfolio win rate would barely move. The risk is overfitting to a tiny sample — 50 trades is insufficient to distinguish a real edge from random variance, espe 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: W7212428131
Cites:
Cited by:
Related to: