Filtering Trades by the Outcomes of Virtual Positions

Summary

The article proposes maintaining a parallel history of virtual positions generated by a trading system. The virtual trades follow the same entry signals and track stop-loss, take-profit, trailing-stop, and signal-based exits, while a separate filter decides whether to place the corresponding real trade. The examples use a MACD zero-line crossing strategy and evaluate filters based on recent virtual trade outcomes.

The tested rules include pausing after consecutive losses, after a stop-loss exit, or when recent trade profitability compares unfavorably with the preceding trade. A results table reports profit or loss, trade counts, and winning and losing trades for several filter settings; outcomes vary, with some variants reducing trades and some changing aggregate results. The evidence is a single reported test, so it does not demonstrate that filtering improves future performance. The approach also relies on virtual execution and history rules that may not reflect the costs and fills of real trading.

Key ideas

• Virtual positions can record hypothetical outcomes while the system decides whether to place real orders.
• The proposed filters use recent virtual trade results and exit types to gate new trades.
• The examples track virtual stop-loss, take-profit, trailing-stop, and signal-driven closures.
• Reported test results differ across filter variants, showing that fewer trades do not automatically mean better results.
• The document provides no basis for assuming these historical filters will generalize to other data or execution conditions.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.