Using Trade History and Indicators to Filter Losing Trades
Summary
The article presents a workflow for investigating a trading strategy’s entries and finding indicator conditions that may help filter losing trades. It starts from a Kalman filter based expert adviser that showed late exits and losses during sideways markets, then proposes saving indicator values when positions open and comparing them with each trade’s eventual result. The workflow covers extending the test period, exporting and parsing the report, assembling trade records, collecting indicator data, and comparing the updated strategy with its baseline.
The example report analysis describes a win share below 34% and average profit 45% higher than average loss, yet an overall losing balance. The article explains parsing the report’s HTML tables and tracking opening time, direction, volumes, commissions, swaps, and profit, then outlines indicator data classes for later analysis. Its main limitation is that the excerpt is incomplete and the reported sample and strategy are specific; selecting filters from historical trades alone does not establish that they will improve out-of-sample performance.
Key ideas
- Save indicator readings at each position entry and relate them to trade outcomes.
- Extend the test period when the initial trade count is too small for useful statistical analysis.
- Parse the tester report into structured trade records, including costs and partial closes.
- Compare candidate indicator conditions with the baseline strategy before adopting them.
- Historical indicators may help identify flat-market losses, but the described results do not establish out-of-sample effectiveness.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.