Midas Grid emulator: how to test your grid settings on real gold ticks

In short: The Midas Grid emulator is a free browser tool. It runs a grid expert advisor's settings day by day through real gold (XAUUSD) ticks since January 2025 and shows profit, drawdown and the minimum capital. It is a model, not a forecast of income.

To test grid settings on gold history you need no MT5 terminal, no history download and no VPS. Open the emulator, set the capital, lot, number of orders and grid step, pick a period and compare two variants on the same ticks.

What can the Midas Grid emulator do?

A free emulator of the Midas grid EA for gold on real Dukascopy ticks since January 2025, with the base updated daily. It has two modes.

• Settings test. Compares one or two variants: capital, lot, number of orders, grid step, basket stop, restart after a blow-up and profit withdrawal. It runs the period day by day at full speed.
• Replay. Plays any trading day tick by tick with the Midas 5.41 panel: M1 to H1 candles, speed from ×1 to ×500, jumps to a basket's start and close, and a basket journal.

A basket is all the open orders of one grid series, which the EA manages as a single position (glossary).

How do you test grid settings on gold history, step by step?

1. Open “Settings test”. Variant A is already set up like Midas 5.41: $1000, lot 0.01, 11 orders, step 1.95, spread 0.2. Keep it as the reference.
2. In variant B change one setting, for example the number of orders or the step. Change one at a time, so you can see what moved the result (what each setting does).
3. Pick a period: a month, half a year, a year or the whole base.
4. Press “Run” and compare the variant cards and the daily balance chart.
5. Open the hardest day in the “Replay” tab and watch how the grid added orders and why the basket closed at a loss.
6. Settings can be passed in the link: /emulator/?capital=750&base=0.01 opens the emulator with $750 capital and a 0.01 base lot.

How do you read the result: profit, drawdown, minimum capital?

• Profit is the result of the period with your settings. Another period gives another result.
• Maximum drawdown is the deepest fall of the account from its peak. Ask whether you could sit through it on your own account.
• Daily balance. On a martingale grid drawdowns tend to be rare but deep, so look at the dips, not the average slope.
• Minimum capital without a blow-up is how much you would have needed at the start for these settings to have no stop-out over this period. It is not computed when a basket stop or profit withdrawal is on. It turns red when it is above the capital you set.

What does a “blow-up” mean in the emulator?

In the model a blow-up is a stop-out: the account cannot carry the open orders and the positions are closed by force (margin call vs stop-out). With “Start over after a blow-up” ticked (it is on in the reference), the account restarts with the same capital and blow-ups are counted. Unticked, the variant stops at the first one. On a live account a blow-up loses most of the deposit, so check the blow-up count before profit.

Why does each variant run on 6 more feeds?

Each variant also runs on 6 slightly different feeds: price ±1 cent per tick and a different execution delay, like different brokers. On a martingale grid the result depends on exactly where the orders land, so the emulator shows the worst and the best case, not one number. If a variant blew up on even one feed, it can do the same at your broker.

How many runs are free, and how do you sign in?

• Without signing in: 5 runs a day, and periods of up to a month stay open after that.
• With a free code from the Telegram bot @MidasGridBot: 25 runs a day.
• With an active Midas license: 50 a day.
• No daily cap: a month of backtests for $25.

The last month of replay is open to everyone, the full history with a code from the bot. To sign in, press “🔑 Site code” in the bot or send /code; on the same device as Telegram, one tap on “📱 Sign in on this device” is enough.

How accurate is it?

The logic is a line-by-line port of our calibrated engine: on the check days baskets and balance match it to the cent. It is still a model: the spread is constant, there is no slippage or requotes, and add-on, averaging and auto-lock are not wired in yet. A broker will give a different result, and the past does not promise income. Grid and martingale trading with leverage carries risk: with too small a deposit a stop-out is possible (Risk, plainly). The grid calculator shows the room to stop-out; protective settings are in how to limit grid risk.

See also: free gold EA backtest on real ticks and how to test a gold EA before buying.

FAQ

What is the Midas Grid emulator and how much does it cost?

It is a free browser tool that runs the Midas grid EA on real gold ticks. Without signing in you get 5 runs a day, with a code from the @MidasGridBot bot 25, with a Midas license 50. A month of backtests, with no daily cap, costs $25.

How do I test grid settings on gold history without MT5?

Open “Settings test”, change one setting in variant B, pick a period and press “Run”. No terminal, history download or VPS is needed: the test runs on our server and the replay in your browser, on a computer or a phone.

Is there an online tester for any grid EA?

No. The emulator models the Midas EA and tests its settings. It does not run other expert advisors.

Why will my broker's result differ from the emulator?

The emulator uses a constant spread, with no slippage or requotes, and add-on, averaging and auto-lock are not wired in yet. Brokers also differ in prices and execution speed, hence the 6 extra feeds.