Gold Sparrow
- Experts
- Version: 1.4
- Mise à jour: 7 juillet 2026
Important: Significant drawdowns are possible during live trading. Please trade only with funds you can afford to lose, and avoid setting the "LotFactor" too high (recommended range: 0.1–0.2).
*Note: The download JSON file for the latest version is not yet available on the server. Please reset the parameters to default and set "UseRecentModel" to `false` before use.*
A new JSON file is scheduled to be deployed once performance under the default settings begins to stagnate. An announcement regarding this will likely be made in the overview section or the comments area.
Until then, you may see the following error:
WebRequest failed. Status code: -1. Error code: 4014.
However, this will not affect operations.
This EA needs 1000 1H bars.
Please set "Max bars in Chart" under MT5's Tools → Options → Charts to a value of 1,000 or higher. Afterward, temporarily turn off "Auto Scroll" (under Charts), then drag the chart to display and load 1,000 past candlesticks. I apologize for the inconvenience.
-----This feature is not yet supported in the latest version-----
This EA allows you to automatically download parameter values—derived from learning recent chart data—by setting the "UseRecentModel" parameter to `true`.
---------------------------------------------------------------
Gold Sparrow is a trading robot developed by training on 15 years of historical XAUUSD chart data.
Please use it on the 1-hour (H1) chart timeframe.
By incorporating multiple channel-based indicators, the EA can easily determine the current price position within the waves of market movement.
Recommended brokers: Brokers using GMT+3 (Daylight Saving Time) / GMT+2 (Standard Time), such as TitanFX.
-----This feature is not yet supported in the latest version-----
■ How to implement the new model --- Go to MT5 > Tools > Options > Expert Advisors and add "https://greentrailns-stack.github.io/MetaTrader/gold_sparrow_model.json" to the list of allowed URLs. After attaching the EA to a chart, a file named `gold_sparrow_model.json` will be downloaded to the "Files" folder within the MT5 "Common" directory. Setting "UseRecentModel" to `true` applies these newly downloaded parameters to trading (note that backtesting is not possible unless the parameters have been downloaded). While the Gold Sparrow EA is running, it will automatically update the file between 01:00 and 02:00 (MT5 server time) on weekdays if a new JSON file is available.
---------------------------------------------------------------
Parameter Description
■ OrderIntervalMinutes --- A filter that prevents the EA from placing a new order until the specified number of minutes has elapsed since the previous order was placed. ■ This EA is programmed to close the trade if it is in profit five hours after the order is placed.
If the position is still showing a loss after five hours, it will close the trade the moment the profit reaches the value calculated by the formula: Lot Size × MinProfit.
For example, if the order size is 0.01 lots and MinProfit is set to 100, the trade will close once a profit of 100 yen (for JPY accounts) or 100 cents (for USD accounts) is achieved. Please adjust the MinProfit setting accordingly if using a Micro account.
The trade will also be closed once the time specified by the MaxHoldingHours parameter has elapsed, even if the position is in a loss.
Note: Backtest results do not guarantee future profits. Please run the EA on a demo account for a period of time to observe its performance before using it with real funds.
"Version: 1.4 Updated: 7 July 2026" with default settings gives excellent results in backtest on XAUUSD with my broker's real ticks since January 1 to July 17, 2026, showing recovery factor 21.7, profit factor 10.8, and win rate 93% (average win $37, average loss -$48, largest loss -$72). There are 6 additional profit trades (and no losses) in the last 10 days, which is effectively a brief forward-test since the version update (and there's no model update for this version yet). Stress-tests on XAUEUR and XAUAUD also give acceptable results (even if understandably much worse than those for XAUUSD).