What an MT5 report can tell you
A report is a record of selected account activity. This analyzer adds accepted monetary postings, separates external cash movements and computes a realized balance path. It cannot prove the file came from a broker, identify whether an account was live or demo, or reproduce missing executions. Its default sample is fabricated teaching data. Use your own complete export to investigate a specific trading period, and reconcile the starting balance and final totals with the original statement before making any decision.
Deals, positions and cash are different records
One position can involve multiple execution deals. Entries, partial exits and reversals do not correspond one-to-one with complete trades, and commissions can be recorded before an exit. If an HTML report contains both Deals and Positions, this tool uses Deals alone to avoid counting the same activity twice. It reports the unit being counted and includes recorded costs instead of presenting an execution count as a trade win rate. A report containing only closed Positions cannot reveal entry-time costs or cash operations omitted from that table.
Deposits and withdrawals change account balance without being trading profit. The performance path excludes classified cash rows, while the account path includes them. Broker-specific adjustments need inspection: a row labelled balance might be a funding movement or a manually posted cost. Convert ambiguous rows into the documented CSV format and classify them explicitly. Blank required amounts are errors rather than assumed zeros; unsupported or duplicate rows are reported for review.
Realized drawdown does not measure floating risk
Drawdown is the fall from the running peak to a later balance. The performance drawdown shown here uses the chosen opening balance plus recorded non-cash outcomes. Its denominator is each running peak, so changing the opening balance changes the percentage. Floating P/L between postings is absent. A strategy could suffer a much larger intratrade equity loss, or violate a daily loss rule, even if its closed balance path looks modest. Do not use this figure as certification that an account passed a broker or prop-firm rule.
A seeded bootstrap asks a limited question
The simulation asks what arithmetic paths arise if the observed monetary postings are drawn again under a selected sampling rule. Each path starts with the same stated balance; recorded amounts are added without scaling position size. A fixed seed makes the draws repeatable. Individual draws assume independent postings. Blocks retain short sequences, but new block boundaries break longer dependencies and edge observations receive less weight. Loss magnitudes or market regimes absent from the observed sample are not generated by this method.
The displayed percentiles summarize simulated paths, not confidence bounds for your future account. A profitable sample can remain favorable under resampling while being unrepresentative of future markets. Missing fees, different lot sizing, changing volatility and margin rules are outside the calculation. Compare several block lengths as a sensitivity check and review the original sequence alongside the simulation. The tool does not label its below-start frequency as a probability of future profit or ruin.
Local processing and export
The analyzer reads your selected file into browser memory and does not send it to an analysis server or persist it in local storage. Clear local data removes it from the tool state. CSV exports contain accepted normalized postings; JSON exports include diagnostics, methods and any current simulation. Files you download remain on your device until you remove them. Share an export only after checking it for private account identifiers or symbols you do not want to disclose.
Format references: MetaTrader 5 table-based trading report and MQL5 deal properties. Platform documentation explains the source record types; the supported parser subset and sampling assumptions are defined above.