LAB #276 NEUTRAL: LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.3% to 33.9% (+1.6pp — E8 Intelligence Research
IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.3% to 33.9% (+1.6pp, n=1271) SAME-WINDOW EFFECT: Applying the Pool density HIGH (>=90) filter to the current live window would have rejected 3 of the 10 trades (EURJPY SHORT, AUDUSD SHORT, GBPUSD SHORT) as they are currently open/rejected. The remaining 7 trades include 2 wins (USDCHF SHORT, EURUSD SHORT) and 3 losses (GBPUSD LONG, USDJPY SHORT, NZDUSD SHORT), yielding a 28.6% win rate — worse than the unfiltered 30% (3 wins out of 10). The +1.6pp backtest improvement does not replicate here, likely due to small sample size and the filter removing a potential winner (AUDUSD SHORT at 0.70029 is still open, could win). PnL delta would be negative or neutral in this narrow window. OVER-TIME PROJECTION: Over 3-6 months, the filter may sustain a modest win-rate improvement of ~1-2pp if pool density correlates with stronger trend alignment, but the n=12 Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Authors
- Andrew Stewart Caldin
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-07-28
- DOI
- https://doi.org/10.5281/zenodo.21637075
- Primary Topic
- Intelligence, Security, War Strategy
- Type
- preprint