Strategy Tester Analysis

Analyse Strategy reviews a MetaTrader 5 Strategy Tester report independently from a code project. It reads the exported workbook, presents statistics and a balance curve, and asks Xen to explain recorded results. It does not run a backtest or modify the EA.

Retain the source version, inputs, symbol, timeframe, dates and modelling choice with each report. These identify what was measured. One configuration's results do not establish how a later source revision or another broker dataset performs.

For export guidance, see Export a Strategy Tester Report.

Upload the workbook

Export the Strategy Tester report in Excel .xlsx format, choose Analyse Strategy and select Upload Excel report. The file must be no larger than 2 MB. The workflow expects the MetaTrader workbook with its deal records; a CSV, HTML report, screenshot or renamed file is not equivalent.

After validation, Xen displays the imported results and automatically begins explaining them. Read warnings before relying on the interpretation. If a workbook is rejected, export a fresh report rather than rearranging columns or changing its extension.

Understand the measures

Read net results, trade counts, profit factor, loss sequences and drawdown together. A high win rate can accompany losses when losing trades are larger than winners. A favourable total can depend on a short period or a few unusually profitable exits, so inspect the pattern behind the headline.

The displayed curve and reconstructed drawdown use available deal balances. They do not establish the full intratrade equity path or floating losses between exits. MetaTrader can calculate other report measures differently. Ask about discrepancies with that distinction in mind rather than assuming similarly named figures must match.

Ask focused follow-up questions

Ask about loss clusters, dependence on a few trades or evidence that another test is needed. Supply context the workbook cannot establish, such as whether parameters were optimised on the same period or whether this is an unseen sample.

Use findings to design another MetaTrader test instead of repeatedly improving the same sample. For a source change, return to the EA project with actual code and a clear request. Import the next report after retesting so the discussion refers to the right version.