Measuring Strategy Profit Concentration and Outlier Dependence
Summary
This article describes an analyzer for judging whether a backtested strategy’s profits are broadly distributed or depend on a few exceptional trades or days. It defines top-N trade shares against net and gross profit, the Gini coefficient across winners, an outlier stress test that removes the largest winning trades, and a daily consistency measure compared with a configurable limit. The analyzer combines these dimensions into a weighted score and grade, while recommending that users inspect the underlying measures as well.
The proposed workflow loads closing-deal dates and results from a CSV file, then calculates and reports the metrics in MetaTrader 5. Illustrations show that identical net profit and win rate can mask very different distributions, and the removal test is explicitly one-sided because losses remain in the sample. The tool is a sensitivity and concentration diagnostic, not proof that future profits will repeat. Results can also depend on how closing deals are recorded, especially when a position has multiple partial exits.
Key ideas
- Headline net profit and win rate do not show whether gains depend on a small number of trades.
- Top-N shares and the Gini coefficient summarize profit concentration using different denominators and samples.
- Removing the largest winners tests dependence on outliers, but is a one-sided stress test rather than neutral resampling.
- Daily aggregation can identify whether a large share of profit came from one day relative to a configured consistency limit.
- The composite grade summarizes several diagnostics, while the individual metrics are needed to understand the source of fragility.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.