Exit logic can be treated as a survival problem, not a fixed rule. Once a position is open, state changes each bar, so the probability of TP before SL should be conditional on survival and updated with time-varying covariates.

A discrete-time competing-risks model fits this: TP and SL are mutually exclusive events, while time-stop creates censoring. A person-period table (one row per trade per bar) supports proper estimation; open-at-time-stop rows are not “losses”.

A native MQL5 implementation is described: no DLLs, no ONNX, no ALGLIB. Two cause-specific hazards are fit via penalised maximum likelihood, converted to cumulative incidence, then to a bar-by-bar hold/close decision while keeping hard TP/SL orders.

Results: out-of-sample log-loss improves 20–22% vs time-only baseline across EURUSD, GBPUSD, USDJPY, XAUUSD, but fixed decision thresholds...

👉 Read | VPS | @mql5dev

MetaTrader 5 sandbox navigation can be implemented via OS dialogs for convenience or via code to enumerate files and directories. The code-driven path becomes necessary when output must be filtered and ordered by attributes such as creation or modification time.

MQL5 provides ArraySort for numeric arrays, but it does not handle strings. A custom bubble-sort style routine can generalize sorting, yet string ordering requires explicit comparison rules.

Using StringCompare alone produces lexicographic ordering, which misplaces numeric strings. Adding length checks fixes numeric strings but can break dictionary word ordering. A revised comparator that switches strategy restores expected results for both cases.

Bubble sort remains O(n²) and is suitable mainly for small datasets or demonstration, not large directory listings.

👉 Read | Freelance | @mql5dev

SCNN reframes multivariate forecasting as an explicit decomposition problem, addressing non-stationarity and regime shifts that make generic neural models brittle and hard to interpret.

The architecture splits each series into long-term, seasonal, short-term, and cross-series co-evolving components, plus a residual. Each component is normalized with additive and multiplicative terms, then handled by a specialized subnetwork matched to its dynamics.

Decomposition and reconstruction are embedded inside the network, enabling information flow between components. A two-branch module adapts parameters online to changing autocorrelation, while structural regularization discourages noise-driven features.

For trading systems, this yields more stable forecasts during distribution shifts and anomalies, with practical computational cost and clearer diagnos...

👉 Read | CodeBase | @mql5dev

This update refactors the MT5 position indicator to support removing stop-loss and take-profit directly from the chart, alongside the existing “close position” interaction.

Core changes focus on decoupling UI elements into a reusable C_ElementsTrade class. C_IndicatorPosition now holds three pointers (open, SL, TP), initializes them to NULL, allocates with new only when needed, and reliably frees memory with delete while resetting pointers to avoid invalid dereferences.

Event handling is hardened: DispatchMessage first validates pointers, then verifies ticket ownership and server-side position existence. Protective levels are created, updated, or destroyed based on whether the server reports a nonzero SL/TP, keeping the chart synchronized with real trade state.

C_ElementsTrade centralizes creation of the price line and an action button, relying on M...

👉 Read | NeuroBook | @mql5dev

Random-walk MCMC methods degrade in high-dimensional models due to small effective step sizes, high autocorrelation, and low throughput. Hamiltonian Monte Carlo addresses this by using gradients of the log target density to generate long, directed trajectories.

HMC augments parameters z with momentum r and defines a Hamiltonian H(z,r)=E(z)+K(r), with E(z)=-log p(z). Leapfrog integration provides reversible, volume-preserving proposals, with Metropolis correction based on energy error.

A 100D correlated normal test case is used, including adaptive tuning of step size epsilon and a diagonal mass matrix. Warm-up is accelerated via MAP initialization using L-BFGS.

Sampling quality is tracked using Rhat, ESS, MCSE, summary stats, and runtime monitoring of log density stability, acceptance ratio, step size, and divergences.

👉 Read | Quotes | @mql5dev

Adaptive MACD for MetaTrader 5 extends the classic MACD by switching from fixed smoothing to a market-dependent blend. It computes a Pearson correlation between close price and bar index over a configurable window, converts it to R², then uses that value to weight two MACD coefficient sets. Higher trend strength increases responsiveness; lower trend strength increases smoothing.

Output is shown in a separate window with a MACD line, a signal EMA, and a color-coded histogram for MACD minus signal. Histogram states cover strong/weak bullish and bearish momentum, based on sign and whether bars are rising or falling. The MACD line color can auto-adjust for light/dark chart themes or be set manually.

Core inputs include R² length (default 20), fast/slow lengths (10/20), signal length (9), and bull/bear colors plus MACD/signal styling. Usable across instrument...

👉 Read | Quotes | @mql5dev

Left-side multi-timeframe and trade setup panel updated for faster at-a-glance monitoring.

Header now shows Live status, Symbol, and the current Bid price on a single top row with real-time updates.

Multi-timeframe structure block covers M1, M5, M15, M30, and H1. Each timeframe reports the current state (BOS, CHoCH, or Swing Formed) and prints the exact breakout price for validation.

Setup details are presented vertically, one field per line to reduce clutter: status (Active/Completed), direction (BUY/SELL), entry, stop loss with pip risk, TP1 (1:1), TP2 (1:2), TP3 (1:3), and net pips captured.

👉 Read | VPS | @mql5dev

Broker Execution Diagnostics is a read-only script for broker specification analysis and execution troubleshooting prior to Expert Advisor configuration or rejected order review. No trades are opened, modified, or closed. The utility reads the active symbol parameters and prints a structured report on the chart and in the Experts log.

Report fields include bid/ask/spread (points), digits and point size, tick size and tick values, contract size and account currencies, volume limits (min/max/step and directional caps), stop and freeze levels, trade and execution modes, supported filling modes and order features, plus terminal/account/EA trading permissions.

Margin and profit estimates are calculated for configurable test volume and price movement using OrderCalcMargin and OrderCalcProfit. Test volume is normalized to the broker’s min/max/step constraints. Us...

👉 Read | Calendar | @mql5dev

Risk Based Position Size Calculator is a chart indicator used to plan trade volume before execution. It adds draggable Entry and Stop lines, measures the distance, and outputs a broker-normalized position size for a selected risk level. It does not place orders or manage open positions.

Risk can be set as a percentage of balance, percentage of equity, or a fixed amount in account currency. The calculation uses tick size and tick value, with an optional round-turn commission per lot. Volume is rounded down to the broker’s volume step to avoid exceeding the configured risk.

The panel shows direction, entry/stop/target, stop distance, risk amount, risk per lot, raw vs normalized volume, reward-to-risk ratio, and validation status. Position sizing remains an estimate due to slippage, gaps, swaps, and execution differences.

👉 Read | NeuroBook | @mql5dev

A price envelope indicator built on the Nadaraya-Watson estimator with Gaussian smoothing produces adaptive upper and lower bands that follow current volatility and direction.

Signals are generated when price closes beyond either band, with colored arrows marking upper or lower crossings. The logic is non-repainting, keeping historical arrows and band values fixed after calculation to support consistent review and testing.

Configuration includes bandwidth control, envelope multiplier, and selectable applied price. Visual settings cover band and arrow colors, with a lightweight chart footprint. Compatible across symbols and timeframes and operates without external dependencies.

Use as an analytical component only; market risk remains and results are not guaranteed.

👉 Read | AlgoBook | @mql5dev

Broker Info Panel MT5 adds a compact on-chart panel for viewing broker and symbol trading specifications without switching between platform dialogs.

Displayed data includes live Bid/Ask, spread in points and price, tick size and tick value, min/max lot, volume step, stop level, and freeze level. This is relevant when validating a new broker, symbol, or account type where contract settings often differ.

The tool is read-only: it does not place, modify, or close orders. Visible fields can be enabled or disabled via inputs, and the refresh interval is configurable. No external DLLs or third-party libraries are required.

Installation: copy BrokerInfoPanel.mq5 to MQL5\Indicators, compile in MetaEditor, then attach to any chart. Values are sourced from broker-provided trading specifications and will vary by account and symbol.

👉 Read | Docs | @mql5dev

Market Session Separator is a chart-only indicator for intraday work that makes session changes explicit without manual time conversion. It renders up to three configurable windows per day using vertical separators, shaded background bands, or both, with optional labels.

Sessions are defined by start/end hours in server time, including overnight ranges where end precedes start. Colors and labels are independent per session, allowing standard Asian/London/New York presets or custom windows such as overlap periods.

Rendering is kept lightweight: no highs/lows or derived values are computed. Session boxes are rebuilt from the chart’s visible price range so they always span the full view while zooming and scrolling.

Operational controls include a rolling days window with automatic pruning, weekend skipping, selectable line style, and clean deinitializat...

👉 Read | Calendar | @mql5dev

Grid and martingale often look profitable in backtests because results are dominated by completed cycles. Profit factor and expected payoff can be positive per cycle while the equity curve trends up.

The failure mode is the same: an unfinished cycle appears when margin, max orders, or trend length breaks assumptions. One large adverse move offsets prior gains and closes the sequence at a net loss.

Simple MT5 EAs with fixed steps (grid) and recursive sizing (martingale) typically reproduce this behavior. Spread and execution costs then push the long-run outcome further negative, even when many cycles finish in profit.

The core math frames both as zero-expectancy without edge on direction, with blow-up risk concentrated in rare tails.

👉 Read | Freelance | @mql5dev

IronHawk Fibonacci Structure Map applies Fibonacci levels only to confirmed swing highs and lows, then validates those levels with higher-timeframe alignment. It calculates key retracement and extension levels and compares them across up to three timeframes.

When multiple Fibonacci references cluster within a volatility-adjusted distance, the tool groups them into a confluence zone using ATR-based rules. Levels include 23.6, 38.2, 50, 61.8, 78.6, 127.2, and 161.8. Zones are context markers, not trade signals, and are intended to focus attention on price reaction.

Additional outputs include basic market-structure context, Premium/Discount around the active swing, and the nearest level with distance from current price. Configuration covers timeframe selection, lookback, swing depth, per-level toggles, ATR period, cluster sizing, minimum overlaps, view ...

👉 Read | CodeBase | @mql5dev

Work described: pattern search in FX data, automated generation of Expert Advisors, and measurement of how long detected patterns remain statistically usable.

Method focus is brute-force coefficient search versus neural-network map search. Both reduce to selecting parameter bytes, but brute force keeps a fixed algorithm and varies a simple coefficient array, while NN varies a larger strategy representation with higher compute and stronger dependence on implementation quality.

Core model uses a bounded multidimensional Taylor polynomial over recent candles, with inputs based on (Close[i]-Open[i]). Positive output maps to buy, negative to sell, with optional sign inversion and a tunable signal threshold.

Optimization is staged: first, coefficient arrays are ranked on next-bar direction metrics (expected payoff or profit factor). Second, full-period ...

👉 Read | AppStore | @mql5dev

A compact Renko context indicator displays the active Renko direction, count of consecutive confirmed bricks, fixed brick size, last confirmed Renko close, and the exact price level where a reversal becomes confirmed. It also reports the remaining distance to that reversal level measured in bricks, with an optional dashed horizontal line on the chart.

The internal model is built from recent tick history. Continuation confirms after price moves one full brick beyond the last Renko close. Reversal is confirmed only after a two-brick move against the current direction: for an UP run, reversal = last close − 2×brick size; for a DOWN run, reversal = last close + 2×brick size.

The goal is to prevent treating minor counter-moves as reversals. No buy/sell signals are generated, and no prediction logic is included. Configuration covers brick size in points, tick...

👉 Read | AppStore | @mql5dev

GDS Renko Bricks renders a fixed-size Renko view directly inside a standard MT5 chart window, without relying on offline charts or custom symbols. Brick formation is driven by tick data, producing completed bricks in real time.

The display keeps uniform brick width and hides the usual time scale, since construction is based on price movement. Reversals follow a two-brick logic consistent with classic Renko rules.

Configuration is minimal: a single Brick Size input. The selected chart timeframe acts only as a host and does not affect Renko generation. Example: on XAUUSD, Brick Size 1.00 produces $1 bricks. No DLLs or external libraries are required.

👉 Read | CodeBase | @mql5dev

System logic uses two internal Renko streams built from BID ticks. A slow stream defines direction, while a fast stream provides the entry trigger. Both apply classic two-brick reversal rules, and entries are permitted only when fast and slow directions match.

Exits are handled by a small target, stop, opposite fast Renko reversal, or a maximum holding time. After closing, a short cooldown in Renko bricks blocks immediate re-entry. Only one position per symbol is allowed, and a spread filter blocks entries when spread is large relative to the fast brick.

Default inputs: Fast 21, Slow 39, Entry Run 1, TP 1.4 fast bricks, SL 2.1, Max Hold 200 minutes, Cooldown 3, Max Spread Fraction 0.57, Lots 0.01.

Example tester run (XAUUSD, M1 host, real ticks, 2026-05-01 to 2026-08-24): 79 trades, +589.34 USD, PF 2.27, max DD 1.18%, win rate 64.56%. Historical demonstrat...

👉 Read | AlgoBook | @mql5dev

An EA implementation that constructs fixed-size Renko bricks directly from BID ticks, using classic two-brick reversal confirmation. A confirmed DOWN->UP reversal defines a support zone; a confirmed UP->DOWN reversal defines a resistance zone.

Entries require displacement: price must move away from the new zone before any retest qualifies. On a return to the zone, a completed Renko brick confirming rejection triggers an order. Each zone is tradable once, with a single active position per symbol. Exits occur via TP, SL, opposite reversal, or maximum holding time. A spread filter blocks trades when spread is large relative to brick size. No DLLs, external libraries, custom indicators, custom symbols, or offline Renko charts; the chart timeframe is only a host.

Defaults: Brick 18.0; Zone half width 1.65; Move away 0.90; Max age 53 bricks; TP 5.50; SL 2.90; Hold 6...

👉 Read | Forum | @mql5dev

This article connects sorting/searching basics to core data structures that make trading code faster and easier to reason about in MQL5: queues, as a foundation for later lists and trees.

It first builds a FIFO queue using a dynamic array with two operations: append at the end and restore from index 0. Removing index 0 triggers a shift, preserving insertion order and providing a clear “empty” return value.

It then upgrades to a circular queue (ring buffer) with fixed capacity, using read/write indices that wrap around. New data overwrites the oldest when full, avoiding costly shifts—ideal for sliding-window indicators like moving averages and other stream-processing tasks in MetaTrader 5.

👉 Read | AlgoBook | @mql5dev