Match the test to the live environment
A backtest is only as useful as its assumptions. Use the same symbol suffix, contract size and spread profile you expect in live trading.
Choose a representative range
Include trending, ranging and high-volatility periods. A test that only covers a quiet quarter will overstate the strategy's robustness.
Read the trade list
- Check the number of trades and the average holding time.
- Look for clusters of losses around news events.
- Compare results across two nearby date ranges.
- Test with a higher spread to see how fragile the edge is.
Keep a baseline
Save the parameters and the report for every build you export, so a later change can be compared against a known baseline.
ThẻbacktestingriskMT5