History of MQL5.community development

The article upgrades SuperTrend by integrating a divergence engine (MPO4 or RSI) the dynamically reduces the ATR multiplier during weakening momentum. It covers the shrinking formula, non-repainting state propagation with dedicated buffers, and a step-by-step MQL5 implementation on the price chart. You will learn how to interpret arrows and line flips, adjust inputs, and apply the indicator for disciplined trailing and earlier confirmations.

A templated CCircularBuffer class for MQL5 replaces the O(n) ArrayCopy array-shift pattern with O(1) insertion using a fixed-capacity ring buffer. The implementation is shown end to end and integrated into a rolling standard deviation indicator. Benchmarks across multiple window sizes compare both approaches and quantify the impact on real-time indicator calculations.

We build an automated MQL5 program that trades Turtle Soup by fading false breakouts of the N-bar high and low. The article implements liquidity-sweep detection, confirmation closes back inside the level, sweep-depth and extreme-age filters, and an optional reversal-candle body check. It adds configurable dynamic or static stops, two take-profit modes, points-based trailing, and clear chart visuals, providing a ready baseline for backtesting and further customization.

This article will illustrate to the reader how to implement a mean-reverting strategy for the EURUSD pair. The strategy follows contrarian trading rules. Our strategy implements a weekly moving average channel, with one moving average on the high-price feed and the latter on the low-price feed. We enter short positions when the price falls beneath the low moving average and long positions when the price rises above the high moving average. Additionally, we will export daily market data to build a simple ONNX model of the market to provide an additional filter for our entries. This provides the reader with a reproducible template for strategy development and backtesting.

Standard MQL5 risk tools read risk from recent history and miss how heavy the downside tail can be. We implement Extreme Value Theory in MetaTrader 5: a Peaks‑Over‑Threshold fit of the Generalized Pareto Distribution via ALGLIB, a live indicator that reports EVT VaR/ES and tail shape, and an EA that sizes positions from the tail estimate. A controlled backtest illustrates reduced drawdown for unchanged entries.

This article demonstrates how to build the UT Bot Alerts indicator in MQL5 using a clear, step-by-step approach. The tutorial explains how to implement an ATR-based trailing stop system, compute a custom EMA for signal detection, and generate buy and sell signals without repainting. The final indicator provides well-structured buffers that enable easy integration with Expert Advisors, automated trading systems, and other algorithmic tools within the MetaTrader 5 platform.

The content we will cover from this point on is much more complex in terms of theory and concepts. I will try to make the material as simple as possible. The programming part itself is quite simple and straightforward. But if you do not understand the theory behind it, you will be left with no practical basis at all for refining or adapting the replay/simulation system to tasks different from the ones I am going to show. I do not want you merely to compile and use the code I present. I want you to learn, understand and, if possible, be able to create something even better.

This is perhaps the most entertaining article so far. The reason is that here we will modify an object already available in MetaTrader 5 in order to create another one that is not originally present on the platform. Of course, what we are going to look at here may seem a little crazy, but it works and serves a very interesting purpose.

This article extends Part 1 by giving an AI access to the development lifecycle on MQL5 Algo Forge. We implement an MCP server over the Forgejo REST API so an agent can create repositories, commit Expert Advisors, branch from main, open pull requests, file issues, and tag releases. You will get a ready-to-run Python server, clear tools, and a safer, reversible workflow.

The article builds a reusable validation layer for Expert Advisors in MQL5. It implements lot-size rules and normalization, SL/TP and freeze-level guards, price digit normalization, margin sufficiency checks, unchanged-level filtering on modifications, account order-limit control, new-bar detection, symbol tradability checks, economic-calendar news windows, and session detectors. The result is cleaner code and fewer terminal errors in live trading.

We add a pinned-tools ribbon: a floating bar that exposes frequently used tools for one-click access without reopening the sidebar. The article implements the ordered pin set and its API, an anti-aliased pushpin control in the flyout, and the ribbon with offscreen clipping, user-resizable width, and horizontal scrolling. The result is faster activation of favorite tools from a draggable, resizable ribbon on the chart.

Parameter optimization inside MetaTrader 5's Strategy Tester routinely produces strategies that perform well in-sample and collapse on forward data. This article builds a native MQL5 Walk-Forward Efficiency scoring engine that quantifies how much of a strategy's in-sample Sharpe ratio transfers to each out-of-sample window. The distribution is rendered as a CCanvas histogram and validated against real EURUSD Daily backtest data.

We implement a symbol resolution framework that abstracts broker naming differences in MetaTrader 5. Using a persistent mapping store, layered resolution with validation, a hash-indexed registry, and a cache, it returns selectable symbols with live market data and logs unresolved cases. Practically, you can deploy the same EA across brokers and keep symbol access consistent at low runtime cost.

We complete persistent homology for MQL5 by reducing the Vietoris–Rips boundary matrix to a persistence diagram. The article implements Z/2 column reduction (CTDAReduction), a diagram container with analytics (CTDADiagram), and a facade that runs the six-stage pipeline in one call (CTDA). Outputs are cross-checked against Ripser to numerical agreement, enabling reliable diagram-based metrics.

This article demonstrates a working prototype integrating Brain-Computer Interface technology with MetaTrader 5, proving thought-based trading is feasible at the software level. A Python Flask server simulates neural command generation, communicating with an MQL5 Expert Advisor via JSON-over-HTTP. The complete pipeline—from signal generation to trade execution—is validated through WebRequest and CTrade. While BCI hardware remains clinically restricted, this simulation establishes a reference architecture for future accessibility options, enabling direct intention-based trading that expands how traders can interact with financial markets.

In this article, we introduce the Mamba4Cast framework and take a closer look at one of its key components: timestamp-based positional encoding. The article shows shows how time embedding is formed taking into account the calendar structure of the data.

The article discusses the adaptation and practical implementation of the ACEFormer framework using MQL5 in the context of algorithmic trading. It presents key architectural decisions, training features, and model testing results on real data.

In today's article, we will see how a simple code change that makes a neuron slightly more specialized can significantly speed up the training stage. After all, once a neuron or neural network, as we will see later, has been trained, the work it performs becomes much faster. We will also discuss a problem that exists but is rarely mentioned.