Frequently asked questions

Any MetaTrader 5 broker that offers the instrument. For MNQ and NQ you need a broker that gives MetaTrader 5 access to CME futures; the backtest and the forward test use AMP's data and accounts. For the NAS100 CFD profile, most CFD brokers offer a Nasdaq-100 CFD under a name such as NAS100, US100 or USTEC; that profile is still in testing.

We are not affiliated with any broker and we do not receive commissions from brokers.

Both. On a netting account (one position per symbol, typical for futures) the expert advisor will not open a trade if another position already exists on the same symbol, for example a manual trade, and it writes why in the journal. On a hedging account (typical for CFD brokers) it tracks its own positions by ticket and still holds one position at a time.

Netting is what the backtest and the forward test use. Hedging has been tested in the Strategy Tester on a hedging demo account, with no errors, but not yet in a live forward test.

Yes. It manages only positions with its own two magic numbers and never touches others. Keep in mind that all programs share the same margin, and that on a netting account a foreign position on the same symbol blocks its new trades until that position is closed.

Hypothetical backtest, not real trading

Our calculation, not investment advice: the largest drawdown in the cautious backtest scenario plus a buffer. For 1 MNQ contract that is $2,954.50 + $500, about $3,500; for NQ, about $35,000. On a NAS100 CFD (still in testing) the position can be as small as $0.10 per point at many brokers, and the balance shrinks in proportion, to a few hundred dollars. On top of that come about $15 a month for market data and $10 to $15 a month for a VPS. Future drawdowns can be larger. Details, margins and the NAS100 CFD figures are on the budget page.

In New York time: Strategy 1 from 18:30 to 08:55 and from 10:55 to 16:55; Strategy 2 after the 09:30 open, with no new entries after 11:00 and an exit by 15:55 at the latest. Positions are closed 5 minutes before the daily CME halt. Details on how it works.

No. The windows are defined in New York time and the expert advisor follows the US daylight saving rules and your broker's clock on its own. The weeks when Europe and the US change their clocks on different dates are handled too. Your computer's own clock and time zone must be correct.

MNQ and NQ are quarterly contracts (March, June, September, December). The expert advisor trades the symbol of the chart it is attached to. Before each roll, when volume moves to the next contract, remove it from the old contract's chart after its position is closed and attach it to the new contract's chart with the same inputs. A good time is the weekend, when it holds no position.

Strategy 1 trades through the overnight session, so a position can be open at night. It never holds a position through the daily CME halt or the weekend break: positions are closed 5 minutes before. Strategy 2 positions are closed the same day.

It is your responsibility to check your firm's rules before you start. Some prop firms do not allow third-party expert advisors, or do not allow an expert advisor that many of their clients run at the same time, because it produces identical trades on many accounts. Rules about news trading, holding positions overnight and minimum trading days also vary.

The inputs include an optional maximum daily loss and minimum equity; when one is reached, the program closes its own positions and stops trading. They are a convenience, not a guarantee: set them below your firm's limits. Most prop firms offer Nasdaq-100 CFDs, and our NAS100 CFD profile is still in testing.

Several reasons, and usually more than one at the same time:

• Costs. Our backtest deducts $2.00 per MNQ round turn. A higher commission, exchange fees or a wider spread lower the result, most of all for Strategy 1, which trades often.
• Price data. Each broker's tick history is slightly different. A small difference in one price can trigger or skip a trade and change what follows.
• CFD versus futures. A CFD follows the cash index, not the futures contract. It has different prices, different overnight spreads and sometimes different session times.
• Test period and contracts. We test each quarterly contract in the period when it was the front month. A test on one contract over a longer period, or on a continuous symbol, will differ.
• Commission inside the deal. At brokers that charge commission inside each deal, a trade that wins by a few dollars can become a loss, which changes the pause after winning trades and the daily target.
• Execution. Slippage, latency, rejected orders and disconnections exist live and not in the tester.
• Settings. A different volume, a different Strategy 1 stop or target, or a strategy switched off gives a different result.

The backtest is hypothetical by definition, but you can reproduce it: the free demo on the MQL5 product page runs in your own Strategy Tester, on your broker's data, and you can see every trade. The forward test will be published as a signal on MQL5, where the account is monitored by MQL5 and not by us, and every trade will be listed on the results page. We do not publish screenshots of balances as proof.

You should not draw conclusions from a few weeks. About 60% of trades lose and the backtest had losing months and a streak of 18 losing trades, so a short period can look much better or much worse than the long run. Every month we compare the forward test with a backtest of the same period. That comparison, over several months, is what tells you whether the backtest can be trusted.

The expert advisor keeps following its rules. It does not increase the volume, add to losing positions or widen stops to recover. In the backtest, the longest streak was 18 losing trades in a row, from 16 to 25 September 2024, worth -$1,376.50 on 1 MNQ contract with costs (hypothetical backtest, not real trading). Decide before you start what loss would make you stop, and set the optional daily loss limit if it helps you.

Several normal reasons: no valid signal inside the trading windows; a position is already open; the daily profit target or the pause after three winning trades was reached; it is a CME early-close day; on a netting account another position exists on the same symbol; or the free margin was not enough. The panel on the chart and the Experts journal show the reason.

The rules and settings were developed during research that used market data from the same period as the backtest, so the backtest is in-sample. That is why we run and publish a forward test on new data. See how the backtest was made.

It will place trades on other symbols, because the MQL5 Market checks every product on forex and gold, but it was designed and tested for the Nasdaq-100 only. Results on other symbols say nothing about the strategy. We do not support it there.

It is the same program with the same rules. The direct subscription costs $49 a month at the launch price, covers one trading account, includes a 14-day refund on the first payment and needs a short licence check over the internet. The MQL5 licence is handled entirely by MQL5, counts computers instead of accounts and has no refunds. Compare them on the pricing page.

You keep access until the end of the period you paid for. After that the program stops opening new trades; positions it already opened keep their stop and target and are managed as usual. We never close your positions at market because a subscription ended.

Yes, up to twice a month: write to us with your licence key and the new account number. One subscription works on one account at a time.

For new subscribers, yes: from $49 to $99 a month once the forward test has three full months of public results, announced on the pricing page at least 14 days before. If you subscribe at $49, you keep $49 as long as your subscription stays active.

No. One position at a time, a stop loss on every trade, and no volume increase after a loss.

No. We publish the behavior (windows, stops, limits), the full backtest with losses and the forward test, but not the entry conditions. You can see every trade it makes in the free demo in your own Strategy Tester.

The MQL5 product page offers a free demo that runs in the Strategy Tester. Rental periods, when offered, are listed there too. MQL5 does not offer refunds, so please test first.

MQL5 links each purchase to a number of activations, one per computer or VPS where you install it. The number is shown on the product page.

Through the MQL5 Market, in MetaTrader 5. Bug fixes are free. The built-in calendar of CME early-close days is updated every December. Every version is listed in the changelog.