How-to guides
How to Backtest SmartEA Gold EAs: Exact Settings, Step by Step
SmartEA Research, 2026-10-28, updated 2026-11-19, 6 min read

Short answer: create a free SmartEA account, download the EA from your dashboard, copy the .ex5 file into MQL5 > Experts, log in to a hedging demo or live account, open the Strategy Tester (Ctrl+R) and run XAUUSD on H1 with Every tick based on real ticks, a $1,000 deposit and leverage 1:500. You need no license and no payment: the Strategy Tester is always free. With the settings below you get a test you can compare directly with the backtest charts on our EA pages.
We publish our backtests, but a backtest you run yourself, with your own broker's prices, is worth more than any screenshot. This guide shows exactly how we test, so you can repeat it.
What you need
- MetaTrader 5 on Windows (or a Windows VPS). Any broker's MT5 works for testing.
- An MT5 account with hedging margin mode. A free demo account is enough. Our EAs run buy and sell strategies side by side and do not start on a netting account, also not in the tester.
- A free account on smartea.trading to download the EA.
Step 1: Download the EA
Create an account (or log in), open Downloads in your dashboard and download the EA you want to test. You get one .ex5 file per EA. The same file is used for backtesting and, once you buy a license, for live trading.
Step 2: Install it in MT5
- In MT5 click File > Open Data Folder.
- Open MQL5 > Experts and paste the .ex5 file there. Do not put it into the Market subfolder: MT5 only loads MQL5 Market purchases from it.
- Back in MT5, right-click Expert Advisors in the Navigator and choose Refresh. The EA now appears in the list.
More detail, including live charts, is in how to install an EA in MT5.
Step 3: Set up the Strategy Tester
Open View > Strategy Tester (Ctrl+R) and choose a single test of an Expert Advisor. Then use these settings, the same ones we use for the backtests on our site:
- Expert: Smart Rider Gold, Smart Striker Gold or Smart Archer Gold (Smart Breaker BTC works the same way on the BTCUSD symbol)
- Symbol: Your broker's gold symbol (XAUUSD, XAUUSD+, XAUUSD.r ...)
- Timeframe: H1 (the EAs read the other timeframes they need by themselves)
- Date: Custom period, for example 2026.01.01 to today
- Modelling: Every tick based on real ticks
- Deposit: 1000 USD
- Leverage: 1:500
- Delays: 300 ms (or a random delay, a slightly harsher test)
- Inputs: Leave the defaults: 0.01 lot per $1,000 of balance, all strategies on, drawdown protection off
No .set file is needed: the defaults are the settings we optimised the EAs with. The news filter and the license check switch themselves off in the tester, so the test only shows the trading logic.
Step 4: Run it
Click Start. The first run with real ticks can take several minutes, because MT5 first downloads the tick history for gold from your broker. Later runs over the same period are much faster.
Tick Visual mode if you want to watch the EA trade on a chart with its panel: open baskets, grid levels, targets and the strategy states. It is slow over long periods, but a few days in visual mode show quickly how the EA works.
Step 5: Compare with our results
Open the Backtest tab when the test has finished. These are our results with exactly the settings above, from 1 January to 26 October 2026:
- Smart Rider Gold EA (version 1.01): end balance $70,405, max drawdown 20.9%, profit factor 5.04, 905 trades
- Smart Striker Gold EA: end balance $11,230, max drawdown 14.2%, profit factor 4.17, 533 trades
- Smart Archer Gold EA: end balance $127,455, max drawdown 26.7%, profit factor 3.19, 2,427 trades
- Smart Breaker BTC EA (BTCUSD, 1 January to 18 November 2026): end balance $7,544, max drawdown 6.3%, profit factor 2.19, 10,577 trades
Max drawdown is the report's Equity Drawdown Relative: the largest drop of equity in percent, open positions included. Because the lot grows with the balance, this is the drop you would have felt at any account size. Look at it before you look at the profit.
The end balances are large because the lot grows with the balance: profits are reinvested every time a basket or trade starts. That compounding also makes later losses larger in money, which is why the drawdown in percent is the number to judge by. The same tests with a fixed 0.01 lot, and the deepest baskets of the year, are in our 2026 backtest results explained.
Why your numbers will differ from ours
A different result does not mean that something is wrong. Your test uses your broker's data, and brokers differ in:
- Prices and spreads. Gold spreads and the tick history are not identical between brokers.
- Server time. The strategies follow server time. Use a broker with the common GMT+2 / GMT+3 server time; on another time zone the sessions shift and the results change a lot.
- History quality. If the broker's real tick history is incomplete, MT5 fills the gaps with generated data. The report shows the history quality.
- Costs. Commission and swap settings of your account type.
Small differences in prices can also change which strategy opens a basket at a given moment, so individual trades can differ even when the overall picture is the same. Compare the shape of the equity curve, the drawdown and the profit factor rather than the exact end balance.
Common problems
- The EA is not in the Expert list: it is not in MQL5 > Experts (or it is in Market), or the Navigator was not refreshed. Restarting MT5 also works.
- The test stops at once with "a HEDGING account is required": the account you are logged in to uses netting. Open a hedging demo account with your broker and log in to it.
- No trades: check that the symbol is gold, the timeframe is H1 and the period is long enough. Some strategies trade only a few times a month.
- The test is very slow: real ticks are heavy. Test one year at a time, and keep visual mode for short periods.
After the backtest
A backtest shows how an EA behaved on past prices. Next, compare it with the live results linked on each EA page and in our Telegram channel, where we post the result of every closed basket, losses included. Then run the EA on a demo account for a few weeks before you trade real money. For a deeper look at the tester and every number in the report, read how to backtest an EA in the MT5 Strategy Tester.
Frequently asked questions
Do I need to buy a license to backtest?
No. Every SmartEA EA runs in the Strategy Tester without a license and without payment. A license is only needed to open trades on a demo or live account.
Can I test other settings?
Yes. Change the inputs in the tester's Inputs tab, for example a larger balance per 0.01 lot for lower risk, or switch on the drawdown protection at the level you would use. Test the exact settings you plan to trade.
Why do you test from January 2026?
It is the period of our published charts and the latest market conditions. You can and should also test earlier years, as far back as your broker's tick history goes.
Backtests are simulations on past prices and do not predict future results. Live trading includes slippage, changing spreads and execution delays. Test on a demo account first and only trade money you can afford to lose.
Follow SmartEA Insider on Telegram
SmartEA Insider: results of our own live accounts, new versions and release notes. No trade signals.


