Gold Order Flow

Gold Order Flow is a rule-based Expert Advisor for XAUUSD (gold). It reads short-term buying and selling pressure from the price ticks your broker sends inside every M15 candle, and follows the market when that pressure becomes unusually strong. One position at a time, with a stop loss and take profit on every order. This is the same EA that trades the Gold Order Flow signal on MQL5, so you can watch it on a real account before you buy.

Who it is for

  • Traders who want a tick-based momentum approach that does not rely on classic indicators
  • Traders who can leave the EA running for about a week of warm-up before the first trade

At a glance

  • Symbol and chart: XAUUSD, M15 chart
  • Trading: any time of day, long and short
  • What is measured: up and down price changes received by the EA from your broker's feed - a tick-based pressure reading, not exchange order flow or traded volume
  • Warm-up: collects tick statistics for about 5 to 6 trading days before the first trade; on live and demo accounts the statistics are saved every candle and reused after a restart
  • Protection: stop loss and take profit on every order and a time limit
  • Backtest averages: about 15 trades per month; median holding time about 25.0 hours
  • Money management: fixed lot (default 0.01) or risk % per trade
  • Gold price filter: new trades only while gold is above 3000 (adjustable input)
  • Status panel: the chart shows the mode, open positions and the reason for the last skipped signal
  • Not used: martingale, grid, averaging down, hedging, DLL, AI

How it works

  1. Tick reading. Every price update received by the EA is classified as buying or selling pressure for the current M15 candle.
  2. Unusual pressure. The recent pressure is compared with the EA's own recent history; the threshold adapts once a day using past data only.
  3. Protected order. The stop loss is a fixed percentage of the price and the take profit a multiple of the stop. The stop loss and the take profit are part of the order request, and the EA checks the stops against the broker's minimum stop distance before sending. If the broker rejects stops in the order request, the EA opens the trade and attaches them right after; a trade that still has no stop loss is closed by the EA. After a connection timeout the EA waits to confirm the order instead of sending it again.
  4. Exit. At the take profit, at the stop loss, or by a time limit.

Because the EA reads the ticks it receives, results depend on your broker's price feed and on the EA running continuously; live results can drift further from the backtest than with candle-based EAs. Saved statistics are used only on the same account, server, symbol and timeframe, and only if they are less than 7 days old.

Setup

  • Attach the EA to an XAUUSD M15 chart and enable AutoTrading.
  • Use a low-spread account (ECN or raw spread) and a VPS. Results depend on your broker's spread, commission, slippage and swap.
  • Default settings (fixed 0.01 lot) are the settings of the published test.
  • Balance-based sizing: set Risk per trade, for example 0.5% or 1%. Percentage sizing targets the planned stop-loss risk at order entry. Signals are skipped when the broker's minimum lot exceeds that budget. Slippage, gaps, fees and overnight costs can cause realised losses to exceed the planned amount.
  • Suggested balance: 2,000 USD or more for each 0.01 lot (largest drawdown in the published test 421.56 USD, about 21% of 2,000 USD). 1,000 USD per 0.01 lot also works but doubles the risk (about 42%).
  • Start on a demo account with the smallest lot.

Input parameters

  • Risk per trade, % of balance - default 0. 0 uses the fixed lot. Above 0 the lot is calculated from the stop distance so that the planned stop-loss risk matches this percentage; signals are skipped when the broker's minimum lot would exceed it.
  • Fixed lot size - default 0.01. Used when the risk is 0. If it is below the broker's minimum lot, the minimum lot is used.
  • Max spread to open a trade - default 1.50. In price units: 1.50 means 1.50 USD on XAUUSD. 0 turns the filter off.
  • Gold (XAUUSD) must be above this price - default 3000. Applies to gold symbols only; 0 turns it off. Open trades continue to be managed.
  • Max slippage, points - default 30.
  • Magic number - default 26111202. Must be above 0. Change it if another EA on the account uses the same number.

Backtest results (MetaTrader 4 Strategy Tester, hypothetical, not live trading)

  • Conditions: Gold Order Flow version 1.0, XAUUSD M15, Every tick, modelling quality 99.90%, variable spread
  • Period: 2025.01.01 - 2026.11.06
  • Deposit and sizing: 10,000 USD, fixed 0.01 lot
  • Trades: 302 (190 long, 112 short)
  • Profit factor: 1.66
  • Net profit: 3,401.24 USD
  • Maximal drawdown: 421.56 USD
  • Average profit / loss trade: 58.46 / -32.91 USD
  • Longest losing streak: 9 trades in a row
  • Commission: not included in the tester result; at 7 USD per lot round turn about 21.14 USD for the 3.02 lots traded

With the same trades, profit and drawdown in USD change roughly in proportion to the lot size; margin, costs and lot rounding can change the result.

Live monitoring

Real account running this strategy (MQL5 Signals): https://www.mql5.com/en/signals/2396982

That account is the author's commercial signal account and trades a larger lot than the default of this EA, so its swings are larger than the published test. Want to watch the account in real time? Send me a private message on MQL5 and I will share read-only (investor) access.

Other charts

Changing to a timeframe other than M15 activates a different, simplified chart-mode strategy: a trend-pullback rule on the chart's own bars, at any time of day, with a stop loss of 1.5 x ATR(14), a take profit of twice the stop and one position at a time; a signal is skipped when the spread is more than half of the stop distance. That strategy is not covered by the published results, and the gold price filter does not apply to it. All recommendations refer to XAUUSD M15.

Testing in the Strategy Tester

  • XAUUSD, M15, model "Every tick", with complete history. Real tick data from your broker gives the most reliable result.
  • Use the default settings to reproduce the published test. The free demo of this product runs in the Strategy Tester, so you can check it on your broker's price data before buying.
  • The "Every tick" model is required because the EA counts ticks. In the tester the statistics start from zero, so the first trade comes after the warm-up.
  • On other timeframes the tester shows the chart-mode strategy described above.

FAQ

  • Why is the EA not opening trades? The status panel on the chart shows the current state and the reason for the last skipped signal, for example: the EA is still collecting tick statistics (warm-up); the price of gold is below the minimum gold price input (3000 by default); the spread is above the limit; the broker's minimum lot would exceed the risk setting (risk % mode); chart history is still loading. Also check that AutoTrading is enabled.
  • Fixed lot or risk %? The default fixed 0.01 lot is the setting of the published test. Risk % sizes each trade from its stop distance. Percentage sizing targets the planned stop-loss risk at order entry. Signals are skipped when the broker's minimum lot exceeds that budget. Slippage, gaps, fees and overnight costs can cause realised losses to exceed the planned amount.
  • Are losing streaks normal? Yes. In the published test the longest losing streak was 9 trades in a row. Size the account so that such a streak is acceptable.
  • Can I run it together with other EAs? Yes, with a different magic number for each EA. Each EA manages only its own trades; the total exposure of several EAs is not limited automatically.
  • What happens on other timeframes? On a chart other than M15 the EA runs a different, simplified chart-mode rule that is not covered by the published results. Use the recommended chart.
  • How do I update the EA? Update when no position is open.
  • Does it use AI? No. It follows fixed, rule-based logic.
  • Why no trade in the first days? The EA first collects tick statistics for about 5 to 6 trading days. The status panel on the chart shows the progress.

Risk warning

Trading leveraged products such as gold CFDs carries a high level of risk and may result in the loss of your entire deposit. Backtest results are hypothetical and do not guarantee future results. Test on a demo account first and trade only with money you can afford to lose.

Support

For questions and help with the setup, send me a private message on MQL5 or write in the Comments tab of this product.

作者のその他のプロダクト
Adaptive SuperTrend
Thanh Trung Tran
インディケータ
Adaptive SuperTrend K-Means is an MT4 port of the open-source TradingView indicator "Machine Learning Adaptive SuperTrend" by AlgoAlpha (Mozilla Public License 2.0). All credit for the idea and the original code goes to AlgoAlpha. At a glance What it is: a SuperTrend whose band width adapts to the current volatility level Volatility levels: K-means sorts recent ATR values into high, medium and low clusters Faithful port: same math and the same defaults as the original Pine Script code For EA dev
FREE
UT Bot Alerts with Signal Stats
Thanh Trung Tran
インディケータ
A line-by-line port of "UT Bot Alerts" by QuantNomad (Pine v4, 2020). The idea is by HPotter and the first version by Yo_adriiiiaan. The original is published as open source on TradingView. All credit goes to those authors. The logic is exactly the original's: ATR is atr(c) = Wilder RMA seeded with an SMA. The built-in iATR is not used. Trailing stop follows the original nested iff / nz recursion: It ratchets up while price stays above the stop. It ratchets down while price stays below the stop.
FREE
Range Filter Buy Sell with Signal Stats
Thanh Trung Tran
インディケータ
A line-by-line port of "Range Filter Buy and Sell 5min" : original Range Filter by DonovanWall simplified version by guikroth Pine v5 update by tvenn It is published as open source on TradingView. All credit goes to those authors. The logic is exactly the original's: Smoothed range = EMA(EMA(|close − close[1]|, period), 2 × period − 1) × multiplier. Each EMA is seeded with an SMA, the same as ta.ema . Range filter: the line stays put until price moves beyond it by more than the smoothed range, t
FREE
Chandelier Exit with Signal Stats
Thanh Trung Tran
インディケータ
A line-by-line port of "Chandelier Exit" , 2019-present Alex Orekhov ( everget ), distributed under the GPL-3.0 license. The original is published on TradingView. This port is released under the same license. The logic is exactly the original's: ATR = multiplier × ta.atr(length) (Wilder RMA, seeded with an SMA). Long stop = highest(close or high, length) − ATR. It only moves up while the previous close stays above the previous stop. Short stop = lowest(close or low, length) + ATR. It moves sym
FREE
Tester vs Live Trade Matcher
Thanh Trung Tran
ユーティリティ
A backtest looks great. Then the EA runs live and the numbers are different, and nobody can say why. Was it slippage, missed signals, a restart or a manual trade? Or was the backtest flattering? TesterVsLive answers that trade by trade. It never opens, modifies or closes orders. What you get: Verdict: green (live matches backtest), amber (partly matches) or red (live differs). Trade match: trades on each side and the match rate (Jaccard) trades the backtest took but live missed, and what they ea
FREE
Gold Priority Ensemble
Thanh Trung Tran
エキスパート
Gold Priority Ensemble combines several rule-based gold signal modules. A fixed priority selects one position per symbol and magic number, with module-specific exit management. Designed and tested for XAUUSD M1. Review the setup, historical test conditions and risk information before testing it with your broker. Who it is for Traders who want one gold EA with several entry ideas but only one open position at a time Traders who accept that trades can last from minutes to several days At a glance
Gold Swing Pullback
Thanh Trung Tran
エキスパート
Gold Swing Pullback is a rule-based gold trading EA designed for an XAUUSD M1 chart. It uses higher-timeframe trend agreement and pullback entries, with stop-based position sizing and break-even management. Its trading style includes frequent small losses and occasional larger trend winners. Historical test results are shown below with their settings and costs; they are not forecasts. Who it is for Traders comfortable with frequent small losses and occasional large trend winners Traders who pref
Gold Inside Day Breakout
Thanh Trung Tran
エキスパート
Gold Inside Day Breakout is a rule-based Expert Advisor for XAUUSD. It waits for an inside day - a day whose whole range stays within the range of the day before - and trades the first breakout of that range on the following day. At most one trade per day, a stop loss on every order and a trailing stop once the trade has moved in its favour. Expect few trades, frequent small losses and occasional larger winners. Historical test results are shown below with their settings and costs; they are not
Gold CCI Breakout
Thanh Trung Tran
エキスパート
Gold CCI Breakout is a fully automated momentum Expert Advisor for XAUUSD. It waits until gold breaks out with real momentum on the H1 chart, then joins the move - long or short. Every order carries a stop loss from the first second, trades are time-limited, and there is never more than one position open. No martingale, no grid. This is the same EA that trades the Gold CCI signal on MQL5, so you can watch it on a real account before you buy. Historical test results are shown below with their set
GoldSwing T24
Thanh Trung Tran
エキスパート
GoldSwing T24 is a fully automated trend-pullback Expert Advisor for XAUUSD. It trades only in the direction of the higher-timeframe trend, enters on pullbacks during the most active hours of the day, and closes every trade within 24 hours at the latest - no position is left open for long. A stop loss and take profit on every order, plus break-even protection after the first move. This is the same EA that trades the GoldSwing T24 signal on MQL5, so you can watch it on a real account before you b
Gold Entry
Thanh Trung Tran
エキスパート
Gold Entry runs two independent trend engines for XAUUSD in one Expert Advisor - one on M30 and one on H1. Each engine reads the market in its own way and manages its own position, so the EA does not depend on a single rule. A stop loss on every order and at most one position per engine. This is the same EA that trades the Gold Entry signal on MQL5, so you can watch it on a real account before you buy. Historical test results are shown below with their settings and costs; they are not forecasts.
GoldIBH
Thanh Trung Tran
エキスパート
GoldIBH combines three complementary gold strategies in one Expert Advisor: a daily range breakout, an H1 momentum strategy and a swing breakout with retest. Each one looks for a different market situation and manages its own position and stop loss, so the EA can find opportunities in more conditions than a single-rule EA. This is the same EA that trades the GoldIBH signal on MQL5, so you can watch it on a real account before you buy. Historical test results are shown below with their settings a
フィルタ:
レビューなし
レビューに返信