Selius · XMQAAlgosLearnTelegram

Learn

How to backtest an MT5 EA on real ticks (and why your results will change)

Step-by-step: set up the MetaTrader 5 Strategy Tester for real-tick backtests, pick realistic spreads and delays, and read the report without fooling yourself.

Why the modelling mode matters

MetaTrader 5's Strategy Tester can simulate price in several ways. Open prices only is fast but only sees one price per bar, so an EA that opens and closes inside a bar is tested against prices it could never have traded. Every tick based on real ticks replays the broker's recorded tick history, including real spreads. For scalpers, breakout EAs and anything with a trailing stop, the difference between the two can be the difference between a smooth curve and a losing system.

Setting it up

  1. Open the Strategy Tester (Ctrl+R) and choose the EA.
  2. Pick the symbol and timeframe the EA was designed for. Check its description or preset name.
  3. Set Modelling to Every tick based on real ticks.
  4. Load the EA's .set file under Inputs → Load.
  5. Set Delays to a random delay, or at least a fixed 50–200 ms, to simulate execution latency.
  6. Use a realistic deposit and your real account currency and leverage.
  7. Run a period that includes at least one volatile stretch and one quiet one.

Reading the report honestly

Skip the net profit and look at maximum drawdown, total trades, the longest losing streak and the recovery factor (net profit ÷ max drawdown). A few dozen trades over several years is an anecdote, not evidence. A profit factor far above 3 on a scalper usually means the test is too optimistic. Then run a second period the EA has never been tuned on: that's your out-of-sample check.

What this proves, and what it doesn't

A real-tick backtest shows how the software behaved on historical data under stated assumptions. It doesn't predict live results. Live trading adds slippage, requotes, VPS latency and changing market regimes. Treat every backtest as the start of the evidence, not the end.

Questions

Is "every tick" the same as real ticks?
No. "Every tick" generates synthetic ticks from minute bars. "Every tick based on real ticks" uses the broker's recorded ticks and spreads, and it's the more realistic of the two.
How long should a backtest be?
Long enough to contain hundreds of trades and more than one market regime. For most intraday EAs that means at least 1–3 years of data.

Put it into practice. Algo Dispensary drops VERUM-screened MT5 EAs every weekday and runs a weekly testing challenge. Members get Vidette, our equity-guard EA, free.

Join Algo DispensaryBrowse the algos

More guides