Evidence before automation
How to evaluate forex robot results
Performance evidence is only useful when you know what generated it. Begin with provenance: vendor backtest, independent reproduction, demo account, or attributable live account.
Risk note: Automated trading can lose money. Historical or simulated performance does not guarantee future results.
Five questions
- What kind of result is this?
- What dates and market regimes does it cover?
- What risk settings produced it?
- How was drawdown calculated?
- Can the result be independently checked or reproduced?
Beware selective presentation
A single attractive account, pair, period or preset may not represent the full range of outcomes. Look for missing settings, discontinued records, deposits/withdrawals that distort growth, and differences between balance and equity risk.