AI MultiModule EA

[yosua_tambunan]

I completely agree with you. Chasing the first breakout blindly on XAUUSD is a recipe for high stress and heavy drawdowns due to institutional stop-hunts.
To solve this exact issue, I developed a hybrid breakout AI algorithm that operates strictly on patience. Instead of jumping on the initial price breach, the EA waits for a cleaner, verified confirmation. It uses an AI Hybrid Breakout logic: it tracks momentum breakout on the RSI indicator but cross-checks it with institutional volume spikes and liquidity sweeps on higher timeframes.
The trade execution is only triggered once a clear candlestick wick rejection forms on the lower timeframe, proving that the fake breakout has ended. Because of this multi-module filter, the EA takes very rare entries (low frequency trading), but it delivers massive profit acceleration with an exceptionally small drawdown—just like a patient sniper waiting for the perfect shot.
If anyone here is interested in testing this low-stress, high-precision approach, I am offering a free 6-month secure trial (.ex4/.ex5) with an optimized .set file. Let me know, and I’d be happy to share it for you to backtest on your own broker data! Let the numbers speak for themselves.

 

[yosua_tambunan]

Sorry, I forgot to include the ReadMe file

 

[Enivid]

WARNING: This expert advisor requires DLL imports to work. Please be careful when enabling DLL imports for closed-source EAs you aren't certain about. This might be dangerous.

 

[yosua_tambunan]

Once again, I forgot to attach the optimal settings for this EA MultiModule_XAUUSD

 

[Enivid]

Once again, I forgot to attach the optimal settings for this EA MultiModule_XAUUSD

Why not make the optimal settings the default ones?

 

[yosua_tambunan]

WARNING: This expert advisor requires DLL imports to work. Please be careful when enabling DLL imports for closed-source EAs you aren't certain about. This might be dangerous.

Thank you to the moderators for highlighting this standard safety warning. I completely understand and support the community's caution regarding closed-source EAs requiring DLL imports.To clear up any concerns, I want to be 100% transparent about why this EA triggers the DLL requirement😀ynamic 6-Month License Locking: Because this is a free, full-feature evaluation version, the code utilizes secure global binary encryption to generate a local license file (.bin) inside the terminal's common data directory. This binds the trial safely to your specific account number.Time-Back Manipulation Protection: The DLL function is strictly used to cross-check your local computer clock with the broker's secure server time. This prevents malicious clock tampering to bypass the trial limit while ensuring your trading parameters remain untouched.This EA performs zero internet scraping, contains no external malware, and does not communicate with any third-party databases. It runs purely as an isolated mathematical sniper engine on your chart.I highly recommend testing this on a Demo Account or within a Sandbox/Virtual Private Server (VPS) environment first. Run your own backtests, verify the near-zero drawdown, and let the secure performance speak for itself!"

Post automatically merged: Jul 27, 2026

Why not make the optimal settings the default ones?

"That is a very fair question! In professional algorithmic development, we intentionally keep the core code parameters at standardized, conservative defaults. This is done to ensure the EA remains a stable 'blank canvas' that can adapt to different brokers, liquidity providers, and server specifications without causing layout conflicts during the initial compilation.
The optimal settings you see in the picture are the result of extensive, rigorous backtesting on high-quality tick data specifically tailored to maximize the current market volatility of XAUUSD. By providing these optimal parameters as a separate preset file, it gives traders the flexibility to load our proven matrix instantly, while still allowing advanced users to tweak the values if their specific broker has different spreads or execution constraints.
Simply apply the values shown in the picture or load the provided .set file in your inputs tab, and you will be running the sniper configuration immediately! Let the numbers speak for themselves."

 

[yosua_tambunan]

Backtest EA_MultiModule_XAUUSD MT5

 

[yosua_tambunan]

Upgrade
Hi everyone,
I would like to share the latest backtest results of my automated trading system, "EA FIB_Kakatua_I_RS". I recently converted it to Pine Script v5 for deep backtesting on TradingView.
This new version is a major upgrade from my previous EA design. In my older version, the strategy frequently suffered from data lag and micro-noise. To fix this, I completely re-engineered the logic into a Multi-Timeframe Execution Matrix to achieve an asymmetrical risk-to-reward ratio.

1. What was Upgraded?

• Macro Trend Filter (H1): Market bias is locked using EMA 50 & 200 on the 1-Hour timeframe to avoid counter-trend trades.
• Liquidity Sweep & Rejection Barrier (H1): It triggers an entry ONLY after a candlestick rejection occurs (minimum 60% wick length).
• AI Adaptive Scaling vs Static Levels: Unlike standard RSI indicators that use rigid 70/30 levels, I integrated an AI Lookback window (144 bars) that automatically rescales the dynamic Fibonacci levels based on actual market volatility over the past week.
• Micro Execution (M15 Execution Upgrade): This is the core upgrade. Once the H1 higher timeframe setup is confirmed, the entry is executed with a tight Stop Loss based on the M15 wick low/high (for scalping-like risk). The exit target is monitored via the real-time M15 RSI.

2. Upgraded Backtest Performance Stats (Jan 2025 - Aug 2026):
I ran the test with a conservative small account setup ($200 USD initial capital) on XAUUSD, and the upgrade delivered phenomenal improvements:

• Total PnL: +$160.75 USD (+80.37% net growth - a massive jump from my previous build)
• Max Drawdown: $21.15 USD (An incredibly safe 5.65% - reduced significantly thanks to the M15 micro-SL upgrade)
• Profit Factor: 1.528 (Upgraded past the critical 1.5 institutional baseline)
• Total Trades: 938 trades
• Win Rate: 20.90% (196 Wins / 742 Losses)

3. Key Takeaway & Analysis:
Even though the win rate sits around 21%, the equity curve climbs consistently. This proves that the upgrade successfully shifted the system into a high-reward, low-risk model. One winning trade easily wipes out 4 to 5 small consecutive micro-losses on the M15 timeframe.
The Run-ups and Drawdowns report confirms that the average profit cycle lasts 12 days (averaging +3.22% growth), while the average drawdown cycle only lasts 7 days (averaging a minimal -1.99% decline).

4. Moving Forward:
The system is originally written in MQL5. Now that the TradingView Pine Script backtest has fully validated this upgrade, I am moving back to MetaTrader 5 to launch it on live micro-accounts.
I would love to hear your thoughts on this Multi-TF upgrade!

1. Do you think increasing the minWickPercent to 70% would improve the win rate further without killing trade frequency too much?
2. How do you handle multi-timeframe synchronization in your own EAs?

Looking forward to a great discussion here. Thanks!

 

[yosua_tambunan]

Test#2