Naive Bayes Gold Bias Pullback: Free MT5 Expert Advisor

Overview

Naive Bayes Gold Bias Pullback primarily targets gold bias and pullbacks but includes logic for universal symbol support. The system uses a machine learning engine that self-retrains on every new bar to predict future price direction. Trade execution depends on detected features like Order Blocks, Fair Value Gaps, and Liquidity Sweeps. Risk management is handled through ATR-adaptive stop losses, daily loss limits, and a volatility gate. The robot also incorporates session-based filters and Murray Math levels.

Self-retraining Naive Bayes machine learning engine

SMC and ICT feature detection including FVG and Order Blocks

ATR-adaptive stop loss and fixed reward-to-risk ratio

Session-based trading filters for London and New York

Gold-specific bias and psychological level analysis

Details

Author

andeleabraham on GitHub

Type

Trading robot

Platform

MetaTrader 5 (MQL5)

Strategy

AI / machine learning, SMC / ICT / price action, Multi-strategy

Markets

Gold, Forex

Timeframes

M15-H1

Risk Profile

No grid, martingale or hedging logic was found in its code or README. That says nothing about how it performs: backtest it first.

EA Files

Source

github.com/andeleabraham/NaiveBayes_EA

Input Parameters

Read from NaiveBayes_GoldBiasPullback_EA.mq5. Defaults are the author's; change them in the EA's Inputs tab.

Name	Type	Default	Description
`LookbackCandles`	`int`	`500`	Training history depth (bars)
`Lookahead`	`int`	`5`	Predict N bars ahead
`RetrainEveryBars`	`int`	`1`	Re-train every N new bars (1=each bar)
`HighProbThreshold`	`double`	`65.0`	Prob% for instant entry
`MedProbThreshold`	`double`	`45.0`	Prob% to arm PREPARING state
`RiskPercent`	`double`	`2.0`	Risk per trade as % of balance
`ATR_SL_Mult`	`double`	`1.5`	SL = ATR × this multiplier
`RR_Ratio`	`double`	`1.8`	TP = SL × RR ratio
`MaxLotSize`	`double`	`0.50`	Hard lot cap
`MaxSpreadPips`	`double`	`3.0`	Block entry above this spread
`ATR_VolatilityGate`	`double`	`3.0`	Min ATR pips to allow trade (filters dead markets)
`AsianStartHour`	`int`	`0`

Show all 40 inputs

Name	Type	Default	Description
`AsianEndHour`	`int`	`8`
`LondonStartHour`	`int`	`8`
`LondonEndHour`	`int`	`16`
`NewYorkStartHour`	`int`	`13`
`NewYorkEndHour`	`int`	`21`
`UseOB`	`bool`	`true`	Order Blocks
`UseFVG`	`bool`	`true`	Fair Value Gaps
`UseBOSCHoCH`	`bool`	`true`	BOS / CHoCH structure
`UseLiquiditySweep`	`bool`	`true`	Liquidity sweep detection
`UseCRT`	`bool`	`true`	Candle Range Theory
`UseSessionAMD`	`bool`	`true`	Session highs/lows (AMD)
`UseH4Bias`	`bool`	`true`	H4/D1 macro trend bias
`UseFibLevels`	`bool`	`true`	Fibonacci retracement levels
`UseMurrayLevels`	`bool`	`true`	Murray Math channel levels
`UsePsychLevels`	`bool`	`true`	Gold psychological levels
`UsePivots`	`bool`	`true`	Daily/Weekly pivots
`OBScanBars`	`int`	`60`	H1 bars to scan for OBs
`OBMinBodyPips`	`double`	`8.0`	Min OB candle body (pips)
`FVGMinGapPips`	`double`	`5.0`	Min gap size (pips) — smaller for M15
`FVGExpiryHours`	`int`	`48`	Hours before FVG expires
`HA_WickTolerancePct`	`double`	`0.10`	Wick tolerance as % of ATR (for metals)
`MaxDailyTrades`	`int`	`8`	Max trades per day (0=unlimited)
`MaxDailyLossUSD`	`double`	`10.0`	Stop trading after daily loss exceeds (0=off)
`BiasBaseLot`	`double`	`0.01`	Normalize gold move value to this lot size
`BiasConfirmUSD`	`double`	`30.0`	Direction is confirmed after this move from day open
`BiasOpenRetestUSD`	`double`	`2.0`	Near day open after confirmation = danger zone
`PullbackSafeMinUSD`	`double`	`20.0`	Safer pullback lower band
`PullbackSafeMaxUSD`	`double`	`30.0`	Safer pullback upper band

How to Install It

1Download the sourceUse "Download source (.zip)" above, or clone andeleabraham/NaiveBayes_EA from GitHub, and unzip it.

2Open the data folderIn MetaTrader choose File, then Open Data Folder, and go to MQL5/Experts.

3Copy the filesCopy the .mq5 file into Experts.

4Compile in MetaEditorOpen the file in MetaEditor and press Compile. Fix any missing includes the compiler reports.

5Test before you tradeRun it in the Strategy Tester, then attach it to a demo chart and allow algorithmic trading.

Third-party code. ForexR did not write, test or endorse this EA, and the summary above was generated automatically from its README and code. ForexR lists only repositories with readable source code and never hosts files. Read the code, backtest it and use a demo account first; automated trading can lose money quickly.

Similar Expert Advisors

My Trading Bot

by carlosrod723

This MetaTrader 5 expert advisor implements a multi-strategy framework based on Smart Money Concepts. It identifies fractal-based liquidity sweeps, order blocks, and Fibonacci zones across multiple timeframes, typically using H4 for trend direction and M15 for execution. The EA includes an optional machine learning fil

MT5 Trading robot SMC / ICT / price actionAI / machine learningMulti-strategy

Updated 10 months ago MIT GitHub

Quantum Alpha MT5 ONNX

by yole-shop-oficial

This expert advisor for MetaTrader 5 combines traditional quantitative strategies with machine learning via the ONNX runtime. It features multiple trading modes including Order Blocks, market structure analysis, and a hybrid AI strategy using neural network predictions trained in PyTorch. The system incorporates compre

MT5 Trading robot SMC / ICT / price actionAI / machine learningTrend following High risk

Updated 5 weeks ago None stated GitHub

RapidPass EA

by jonahgrigoryan

RapidPass EA (RPEA) is an MT5 trading robot designed specifically for passing prop firm challenges. It utilizes a hybrid ensemble architecture combining Burst-Weighted Imbalance with Session Confluence (BWISC) and mean-reversion strategies. The EA incorporates machine learning through Q-learning and contextual bandits

MT5 Trading robot SMC / ICT / price actionMean reversionAI / machine learning High risk

Updated 6 months ago None stated GitHub

Gold Trader EA

by mehdi-jahani

141

Gold Trader EA is a multi-strategy expert advisor designed for trading XAUUSD on the MetaTrader 5 platform. It employs a fusion engine that combines up to 14 different technical analysis methods, including candlestick patterns, chart patterns, price action, and institutional levels. The system requires a user-defined n

MT5 Trading robot Multi-strategyTrend followingIndicator crossover

Updated 1 year ago None stated GitHub

SMC Unified Expert Advisor

by KVignesh122

This MetaTrader 5 expert advisor implements Smart Money Concepts including Order Blocks, Fair Value Gaps, and Break of Structure entries. It features an automated mode that combines multiple strategies alongside market regime detection using ADX or Efficiency Ratio to ensure trading in trending conditions. Risk managem

MT5 Trading robot SMC / ICT / price actionTrend followingMulti-strategy

Updated 2 months ago Apache-2.0 GitHub

Makets GA

by nopaixx

Makets GA is a neural network trading system for MetaTrader 5 that utilizes genetic algorithms to optimize weights for multi-layer perceptrons. Instead of standard price prediction, the EA evolves trading strategies by processing price data across several major currency pairs simultaneously. It implements evolutionary

MT5 Trading robot AI / machine learningMulti-strategy

Updated 8 years ago None stated GitHub