Skip to content
realtrademaster

How to Set Up Forward Testing in MT5 Strategy Tester

The Forward setting in MT5 Strategy Tester lets you check a parameter set on later historical data. The tester searches the earlier portion of the selected dates and runs selected results on the later portion. Choose the split and decide how you will judge the results before you start. That gives you a record of what the later period actually tested.

Steps to check

  1. In Strategy Tester, select the Expert Advisor, broker symbol, timeframe and date range. Check that enough price history is available around your start date. Write down the data modelling mode, execution delay, initial deposit and trading cost assumptions so you can reproduce the run.
  2. Enable optimization and pick the parameters and ranges to search. Set Forward to a half, a third, a quarter or a custom start date. Record the date shown and keep the earlier interval for the search. Decide how many observations and what account equity decline you will consider before you inspect any later results.
  3. Run the test and save the Optimization Results and Forward Results. Record the search method and ranking criterion. MT5 selects forward candidates from the earlier optimization results, with selection rules that depend on the search method and number of passes. Use the actual parameter values to match results across the tabs.
  4. Compare the matched runs on trade count, net outcome after the configured costs, and equity drawdown. Note the length and coverage of each period. A shorter forward window can produce smaller totals even when individual trade outcomes are similar. Inspect trade-level records when you need to understand a difference.
  5. Keep the original settings and both result sets. If you change a range or choose new parameters after reviewing the forward results, record that decision and treat those dates as development data. Use a later untouched interval or a subsequent demo observation to evaluate the revised rules.

Worked example

Suppose you choose January through December and set a custom forward start at the beginning of October. Search the first nine months and compare selected parameter sets on October through December. If a candidate has only four trades in the later window, report the trade count and hold any conclusion that requires a larger sample.

Frequently asked questions

Which part of the dates does MT5 use for the forward period?

It uses the later part of the dates you selected. With a custom split, the date you enter sets the start of that later part.

Where do I compare the same parameter set?

Check the Optimization Results and Forward Results tabs and match the actual inputs. Rankings can change between the earlier and later periods.

Can I compare the forward run directly with live execution?

Keep the historical result and any live observation as separate records. Live orders face actual quotes and broker responses at the time of submission.

Related guides

Related video

Related example: costs to hold consistent when comparing test periods

Watch on YouTube

Sources