Scale Out Value Analyzer - MetaTrader 5 için komut dosyası

Scaling out of a position is usually assumed to help, but no standard MetaTrader report compares the blended result against what the same position would have returned if the trader's own exit prices had been used differently.

The script reads a CSV of closing deals and prints a full report to the Experts tab:

• Value-Add Ratio — the aggregate value scaling added, against holding the full position to the last exit.
• Scale-Out Win Rate — the share of scaled positions where scaling out beat holding to the last exit.
• Scaling Efficiency — the share of the best outcome available among the trader's own exit prices that was actually captured.
• A composite score (A+ to F) combining the three dimensions, with a single-trade dependence check and written recommendations.

Input. Place a CSV named in the InpFileName input (default ScaleOutData.csv) in the terminal MQL5\Files folder, with the header PositionID,Symbol,Direction,EntryTime,EntryPrice,EntryCount,TotalVolume,ExitTime,ExitPrice,ExitVolume,ExitProfit,ExitCommission,ExitSwap and one row per closing deal. On the first run, if no file is found, the script generates a built-in demonstration set and analyzes it, so the report is visible immediately. The companion script ScaleOutExport.mq5 writes that file directly from the terminal's own account history: run it once (optionally setting a from/to date range), then run ScaleOutValue.mq5 on the file it produces.

Parameters. InpFileName sets the input CSV; InpMinScaledPositions sets the minimum number of scaled positions required before a grade is produced; InpWeightValueAdd, InpWeightHitRate and InpWeightEfficiency set the three composite weights (normalized automatically); InpValueAddSensitivity sets how strongly the Value-Add Ratio moves its sub-score; InpGradeAPlus through InpGradeD set the letter-grade boundaries; InpHitRateAlertBelow, InpEfficiencyAlertBelow and InpStressDependenceAlert set the thresholds that trigger each written recommendation.

The tool is written in pure MQL5. It needs no external libraries, no Python and no DLLs. A full description of the method and the code is available in the article: [ARTICLE_URL].