Myrto Asian Range Breakout
- エキスパート
- バージョン: 1.1
- アップデート済み: 6 11月 2026
- アクティベーション: 5
Myrto marks the overnight price range and places stop orders on both sides of it. It was designed for XAUUSD on M15. Every trade has a stop loss, and open trades are closed at a set hour each day.
What it does
Myrto measures the high and low of a daily time window, 00:00 to 07:00 server time by default. After the window ends, it places a buy stop above the range and a sell stop below it. When one order fills, the other is cancelled.
The stop loss sits on the opposite side of the range and the target is 2 times the risk. By default the stop moves close to break-even at 1 R. Pending orders are cancelled at 16:00 and open trades are closed at 21:00, or 19:00 on Friday.
There are at most 2 trades per day. A second trade is allowed only after a loss.
A day is skipped when the range is smaller than 0.25 or larger than 1.0 times the daily ATR.
Myrto does not use a grid, martingale or averaging. It holds one position at a time on its symbol.
Before you use it
Check your broker's server time. All hours are broker server time. The defaults were built for a broker on UTC+2 in winter and UTC+3 in summer. On a broker in another time zone, the range window will cover different hours unless you change the session inputs.
Use a hedging account with XAUUSD or your broker's gold symbol. It was developed and tested on a hedging account and has not been tested on a netting account. It can run on other symbols, but its default settings were developed for gold only and results on other symbols were not studied.
Use one chart per symbol. The chart timeframe does not change the logic. M15 is recommended so that the levels are easy to read.
Main settings
| Input | Default | Meaning |
| RangeStartHour, RangeEndHour | 0, 7 | The range window, in server time |
| LastEntryHour | 16 | No new orders from this hour |
| CancelPendingHour | 16 | Unfilled orders are cancelled |
| CloseHour, FridayCloseHour | 21, 19 | Open trades are closed |
| UseFridayCutoff, FridayLastEntryHour | false, 14 | Optional earlier last entry on Friday |
| LotMode | Fixed | Fixed lot or percent risk |
| FixedLot | 0.01 | Lot size in Fixed mode |
| RiskPct, MaxLot | 1.0, 0.10 | Risk per trade and lot cap in percent mode |
| DailyLossPct | 2.0 | Closes everything for the day at this equity loss |
| EntryBufferPts | 20 | Distance from the range to each order |
| RR | 2.0 | Target as a multiple of the risk |
| TrendMode, TrendEMA | Soft, 200 | H1 trend filter |
| MinRangeATR, MaxRangeATR | 0.25, 1.0 | Range filter, in daily ATR |
| MaxSpreadPts | 350 | No orders above this spread |
| UseBreakEven, BE_TriggerR | true, 1.0 | Move the stop when the trade reaches this R |
| AllowReentry, MaxTradesPerDay | true, 2 | Second trade after a loss |
| Magic | 260120 | Identifies this expert's trades |
The dashboard shows the session state, the range, risk limits and closed trade statistics. It has Pause, Close All and Minimize buttons.
Fixed mode note. The drawdown control, the half size on counter-trend trades and the equity curve scaling all work by changing the percent risk. In Fixed mode, the default, the lot stays at FixedLot and these controls do not change the size. Use percent mode if you want them to act.
Minimum balance for percent mode. At 1% risk on gold with a 0.01 minimum lot, percent mode needs a balance of about 3,000 USD. Below that the lot would risk too much, so the expert skips the order and writes a line to the log. At about 3,000 USD it skips many orders; a 17-month test on one demo server skipped 39% of them, against 1% at 10,000 USD. Fixed mode has no such limit: check the risk of your fixed lot yourself.
Historical test results
These are results of a historical test in the MetaTrader 5 Strategy Tester. They are not live results and they do not show what will happen in the future.
The figures below, including the maximum drawdown, come from one full 2-year test run by the author.
| Item | Value |
| Symbol and timeframe | XAUUSD M15 |
| Period | 2024.10.01 to 2026.09.19 |
| Deposit and leverage | 3,000 EUR, 1:100 |
| Lot | Fixed 0.01 |
| Modelling | Generated every tick, history quality 99% (not real ticks) |
| Net profit | 2,032.60 EUR |
| Profit factor | 1.60 |
| Maximum equity drawdown | 345.72 EUR (8.97%) |
| Trades | 409 |
| Profitable trades | 57.46% |
| Largest losing trade | -75.62 EUR |
| Longest losing run | 6 trades, -137.29 EUR |
This test used the default inputs of this product. Results will differ with your broker's data.
Results by year in this test:
| Year | Trades | Net (EUR) | Profit factor |
| 2024 (from October) | 52 | 19.28 | 1.08 |
| 2025 | 215 | 558.71 | 1.35 |
| 2026 (to September) | 142 | 1,454.61 | 1.95 |
About 72% of the net result came from 2026. Gold was unusually volatile and rising strongly over this period.
An earlier period lost money. A real-tick test of the default settings from 2022.10.01 to 2024.10.09, with a 3,000 USD deposit, ended at -235.88 USD, profit factor 0.86, over 340 trades, with a maximum equity drawdown of 10.81% and a longest losing run of 12 trades. The strategy did better in high volatility and worse in quieter conditions. Please test it yourself on your broker's data.
Risks
- Trading leveraged products can lose all of your deposit. Use only money you can afford to lose.
- Past and tested results do not predict future results. This product makes no promise of profit.
- The settings were chosen on recent gold data, so the historical result may be better than what you get in a new period.
- Losing runs happen. The 2024-26 test had 6 losses in a row and the 2022-24 test had 12.
- Spread, slippage, swap and execution differ between brokers and from the tester.
- News can move price through stop orders, so a loss can be larger than planned.
- The daily loss limit uses total account equity. Other experts or manual trades on the same account can trigger it, and it will then close this expert's trades.
- The expert must run continuously, so the terminal needs a stable connection.
- This product is not financial advice.
Please run it on a demo account first.
Support
Please use the product comments or MQL5 messages.
