Source noteMar 10, 20262 min read

Update · September 28, 2026

A stricter retest, which keeps the engine's trades across many coins together the way it actually trades, found these results no better than chance, and the engine lost money on 2026 data. The numbers below are the 2024-2025 test as first published, kept as that record.

Walk-Forward Performance, 2024-2025

Two years of out-of-sample outcomes with actual Binance funding costs applied per entry. 26 wins out of 29 outcomes. Weak spots left in.

System design

Higher-timeframe regime filter to avoid entries against the macro trend.
Adaptive band system for entry timing.
Multi-day confluence window so isolated signals do not trigger entries alone.

Walk-forward protocol

Train window: 2021-01-01 to 2023-12-31.
Test window: 2024-01-01 to 2025-12-31.
Real Binance historical funding rates applied per entry.
Ten-asset universe with no exclusions.

Primary results

26 wins out of 29 out-of-sample outcomes.
120% cumulative return in the test window.
Average funding drag of roughly 1% per entry.
Five validation tests pass and three fail, all published.

Known limitations

The out-of-sample outcome count is still small.
A few large winners drive most of the P&L.
Performance is regime-dependent: tested across bull and range-bound conditions in 2024-2025, not prolonged bear markets.