How to backtest an EA in MT5

1. 1

   Open the Strategy Tester

   In MT5 press Ctrl+R, or choose View → Strategy Tester. The EA must already be installed and compiled.
2. 2

   Pick the EA, symbol and timeframe

   Choose your Expert Advisor, the symbol it trades, and the timeframe it was designed for.
3. 3

   Set the date range

   Use a custom period long enough to include different market conditions: trends, ranges, and volatile news weeks, not just a recent stretch that happens to suit the strategy.
4. 4

   Choose the modelling mode

   This decides how price moves inside each bar; see the section below.
5. 5

   Set deposit, leverage and inputs

   Use the deposit and leverage you'd really trade with, and check the EA's settings on the Inputs tab. Tick Visual mode if you want to watch the trades being placed.
6. 6

   Start and read the report

   Click Start. When it finishes, the Backtest tab has the report, the Graph tab shows the balance curve, and the Journal lists every order and any errors.

1. 1

   Every tick based on real ticks

   Replays your broker's recorded ticks, including real spreads. The most realistic, the slowest, and only as good as the tick history your broker provides.
2. 2

   Every tick

   Generates artificial ticks inside each one-minute bar. Detailed, but the in-bar path is simulated rather than recorded.
3. 3

   1 minute OHLC

   Uses just four prices per minute: the open, then the low and high (low first on a rising minute, high first on a falling one), then the close. Fast and a good middle ground for EAs that make decisions on closed bars; less exact for tight stops, which can be hit inside a minute in ways four prices don't capture.
4. 4

   Open prices only

   Only each bar's open price. Very fast, but valid only for EAs that act solely at the start of a bar. Stops and trailing stops are modelled crudely.

1. 1

   Total net profit

   Profit after the whole test. Meaningless on its own without the drawdown next to it.
2. 2

   Maximal drawdown

   The largest fall from a peak. This is the pain you'd have had to sit through, and the number prop firm rules are measured against.
3. 3

   Profit factor

   Gross profit divided by gross loss. Above 1 means winners outweighed losers over this period.
4. 4

   Total trades

   Too few trades (a few dozen) and the other numbers are mostly luck.

A backtest shows how your rules behaved on one stretch of history, not what they'll do next. Tuning inputs until the curve looks perfect usually fits noise rather than finding an edge, so test on a period you didn't tune on, then run the EA on a demo account. More on backtesting pitfalls.