Test Any Gold EA in 3 Minutes Before You Pay for It

Every gold EA looks flawless in a screenshot: smooth equity, big numbers, bold promises. Then you run it live and discover what the picture hid—no stop-loss, position sizes that balloon after a loss, trades stuck for days, and one sharp move that wipes the account. A screenshot is not proof. Proof is a test you run yourself, on your broker’s data, before spending a cent. Here’s how to do it in about three minutes using the free demo of EVA Gold Engine in MetaTrader 5.

Step 1: Install the free demo

• Open MetaTrader 5 and go to the Market tab.
• Search for “EVA Gold Engine” and click Free Demo to install.
• No API keys or external registrations are required.
• The demo uses the same engines as the live version, so the tester reflects the live logic you’d run on a chart.

Step 2: Configure the Strategy Tester

• Expert: EVA Gold Engine
• Symbol: XAUUSD (or your broker’s gold symbol; GOLD and suffixed names are recognized)
• Period: M5
• Modelling: Every tick based on real ticks (uses your broker’s real spread and tick data)
• Dates: last 12 months
• Deposit: set what you typically trade
• Inputs: leave defaults (by default, risk per trade is 1% of balance, so results scale with deposit)

Press Start. That’s the entire setup.

Step 3: Watch a single day in Visual mode

Before reading any totals, enable Visual mode and watch one session. Around 07:00 GMT, EVA measures the overnight range of gold and places two pending orders—one above, one below—each with a predefined stop-loss and take-profit. When one order triggers, the other is canceled automatically. You can see the rules unfold trade by trade, candle by candle.

Step 4: Read the report like a skeptic

Ignore net profit at first. Start with the risk lines that expose fragile systems:

• Largest losing trade: With a real stop on every trade, the worst loss should align with the stop size, not balloon unpredictably.
• Lot sizes: They should not increase after losses. If sizing rises as losses pile up, that’s martingale—walk away.
• Time in market: Positions should resolve in hours, not linger for days.
• Maximum drawdown: This shows the worst pain you’d have had to sit through. Make sure it matches your tolerance.

Then rerun the same test using a second broker’s data. You’ll see how much result differences come from spread and execution alone—the silent variable most sellers gloss over.

What the tester shows—and what it doesn’t

In the tester, you get the trading engines, the account guard, and the control panel exactly as they run on a chart. Two live-only conveniences don’t activate in backtests:

• Ask EVA: type a question and get answers about your account in your language (requires a live chart).
• Telegram controls: receive alerts and pause engines from your phone (also live only).

The tester never calls any AI services, which is why you don’t need keys or extra setup to try it.

Three minutes, zero pressure

You wouldn’t hand your account to a stranger without seeing how they trade. Don’t hand it to an EA either. Install the free demo, run the last twelve months on your broker’s data, watch a day in Visual mode, and decide with your own numbers in front of you.

Important: Backtests are not guarantees of future performance. Always validate in a demo or with minimal risk on a live account, and use sizing that fits your risk tolerance.