History of MQL5.community development

The article adds a self-adaptive layer to the Malaysian Engulfing indicator by optimizing the retest bar range with a constrained brute-force search scored by MFE and MAE. It details the data model, helper routines, and an MQL5 implementation that gathers historical setups, computes excursions, and selects the best parameter. Readers learn how to remove manual tuning and run the indicator with context-appropriate settings across symbols and timeframes.

This article extends a Flask backend to reliably receive, validate, and store closed trade data from MetaTrader 5 using SQLite and Flask‑SQLAlchemy. It implements required‑field checks, timestamp conversion, transaction‑safe persistence, and working retrieval endpoints for all trades and single records, plus a basic summary. The result is a complete data pipeline with local testing that records trades and exposes them through a structured API for further analysis.

Learn how to build a manual backtesting EA for MetaTrader 5's visual tester by adding chart buttons with CButton, executing orders through CTrade, and filtering positions with a magic number. The article implements Buy/Sell and Close All controls, configurable lot size and initial SL, and a trailing stop via CPositionInfo. You will also see how to load indicators with tester.tpl to validate ideas faster before automation and narrow optimization ranges.

Biogeography-Based Optimization (BBO) is an elegant global optimization method inspired by natural processes of species migration between islands within archipelagos. The algorithm is based on a simple yet powerful idea: high-quality solutions actively share their characteristics, while low-quality ones actively adopt new features, creating a natural flow of information from the best solutions to the worst. A unique adaptive mutation operator provides an excellent balance between exploration and exploitation. BBO demonstrates high efficiency on a variety of tasks.

The article presents two systematic pitfalls in MQL5 multi‑timeframe work: indicator handle leaks that exhausted resources and repainting from reading the forming bar (index 0). It introduces MTFEngine.mqh, a unified include that creates and tracks handles in one place and defaults all reads to closed bars (index 1). A D1–H4–H1 example shows how this approach keeps signals technically correct and consistent with charts.

The article describes the arrangement of a coordinated ML pipeline in MetaTrader 5 with separation of roles: Python trains and exports the model to ONNX, MQL5 reproduces normalization and PCA via matrix/vector and performs inference. This approach makes the model's inputs stable and verifiable, and the MetaTrader 5 strategy tester provides metrics for analyzing the system behavior.

We will review the basics of Gaussian processes (GP) as a probabilistic machine learning model and demonstrate its application to regression problems using synthetic data.

The article addresses the loss of temporal information in ML pipelines by encoding periodic time variables with Fourier harmonics and adding forex session structure. It implements session and overlap flags, lagged session volatility, and calendar effects, then prunes features by timeframe. The get time features function returns an index‑aligned, ML‑ready set of time features suitable for integration with price‑based signals.

A custom forward simulation engine detects fast/slow EMA crossovers and immediately projects synthetic candles ahead of the signal bar. It generates bodies and wicks using controlled logic, draws them with chart objects, and refreshes on every new signal or anchor change. You get a clear forward-looking view to test timing, visualize scenarios, and manage invalidation on the chart.

We present a rule‑based alphabet for candlestick price action that maps measurable shape and direction to letter codes (A/a, H/h, E/e, G/g, D). The article shows an MQL5 implementation: classifying candles, building two‑bar sequences via permutations, and scanning charts with an indicator and alerts. Readers gain a practical template for objective pattern detection and systematic testing.

In this article, we will consider conformal predictions and the MAPIE library that implements them. This approach is one of the most modern ones in machine learning and allows us to focus on risk management for existing diverse machine learning models. Conformal predictions, by themselves, are not a way to find patterns in data. They only determine the degree of confidence of existing models in predicting specific examples and allow filtering for reliable predictions.

Implement the Malaysian Engulfing concept in MQL5 with two coordinated indicators. One applies strict, body‑based engulfing rules for precise pattern detection; the other uses a state-driven model to monitor what follows—pullbacks and timed retests—directly on the chart. The result is a repeatable, rule-based workflow that replaces visual guesswork with programmable logic.

Learn how to implement a tick-based chart in MQL5 where each bar is built from a fixed number of ticks instead of time. The article covers creating and configuring a custom symbol, capturing real-time ticks, forming OHLC values, and pushing data with CustomRatesUpdate. This approach produces activity-driven candles that better reflect market intensity and short-term momentum for precise intraday analysis.

Learn to engineer an MQL5 indicator that converts trend, momentum, and volatility into a single raw score using a matrix.mqh (ALGLIB). The article covers a separate‑window oscillator to validate the core mathematics, then a main‑chart indicator that plots non‑repainting buy/sell arrows when the score crosses user‑defined thresholds. An optional long‑term EMA filter, a minimum‑bar cooldown, and built‑in alerts make the tool practical for live trading.

This article applies the Optimal Trading Rule from AFML Chapter 13 to set profit targets and stop-losses without in-sample calibration. We model post-entry P&L with a discrete Ornstein–Uhlenbeck process, run a 100,000-path search, and implement Python, multiprocessing, and a Numba @njit parallel kernel (242× faster). The result is an optimal (PT, SL) under three forecast specifications, constrained by the prop-firm daily loss limit.

The article is dedicated to the event-driven architecture in MQL5 and describes the transition from the monolithic OnTick model to distributed processing. We will consider predefined and custom events, services and messaging between programs, as well as common architectural errors. A practical example demonstrates how to organize interactions between indicators and an EA to reduce load, improve readability, and simplify maintenance.