Extralonger targets a common pain point in market forecasting: most models handle minutes or hours, but fail on multi-day horizons. It treats market data as spatiotemporal, aiming to stay stable over long histories while keeping compute and memory costs low.

The core idea is a unified space-time representation that merges price “space” and time into one encoding, avoiding duplicated passes across both dimensions. Architecture-wise, it runs three parallel routes (temporal, spatial, mixed) and uses a Global-Local Spatial Transformer to capture both distant cross-asset links and short-range session-level interactions, with a full receptive field over history.

The MQL5 implementation centers on CNeuronExtralonger, adding adaptive attention pooling so route weights are computed per input, not fixed. Modules include shared temporal projection, separate ...

👉 Read | Docs | @mql5dev

A consensus of LLM agents tends to average away rare but valuable signals and stays static even when specific agents underperform. This article replaces voting with a market-style selection mechanism: each agent trades independently and is judged by PnL, not agreement.

The framework runs 10 distinct trading “philosophies” in parallel, tracks per-agent capital and streaks, and rebuilds each system prompt every cycle via build_prompt() to reflect risk posture. Results persist across sessions, and closed trades trigger a REWARD protocol that updates agent state without extra model calls.

On MT5, true separation is achieved with per-agent magic numbers, allowing simultaneous long/short positions on the same symbol. Three modes cover research and production: observer, full parallel trading, and selection that disables bankrupt agents.

Backtest (EURUSD M15) sho...

👉 Read | AlgoBook | @mql5dev

Most spread indicators average over ticks, which underweights quiet minutes and smooths out the spreads paid during rollover, pre-open, and low-liquidity sessions. A timer-based sampler avoids that bias by taking one spread snapshot per second so each minute has equal weight.

PropSpread is built as a monitoring utility for prop firm and funded accounts. It shows live spread in broker points, as raw price difference, and in the symbol’s natural unit derived from digits and calculation mode (Forex pips, index points for CFD indices, otherwise price units). A legend is always printed to verify conversions.

The panel reports rolling stats over a configurable window (min/avg/median/max and time above threshold), plus session stats since server midnight. It adds spread as a percent of ATR from the last closed bar, and includes threshold alerts with hold time...

👉 Read | Freelance | @mql5dev

EMA Cross Demo EA for MetaTrader 5 was released as a compact reference Expert Advisor focused on development method rather than trading advantage. Signals are evaluated once per new bar using closed candles only, preventing repainting. A fast/slow EMA cross triggers entry; an opposite signal closes the current position and opens the new one on the next tick of the new bar.

Risk is based on ATR: stop-loss uses ATR multiplied by a configurable factor, take-profit is derived from the stop distance via a reward:risk input. The EA enforces the broker’s minimum stop and freeze distances, normalizes volume to symbol limits and step size, and skips entries when spread exceeds a configured maximum.

Position handling is safe for both hedging and netting accounts via magic number and symbol lookup. Recent fixes include correct stop-distance flooring against Bid/Ask on...

👉 Read | VPS | @mql5dev

ATR Trailing Stop ported from TradingView Pine Script to MQL5, kept consistent with a Python reference implementation and validated bar-by-bar.

Rules across all versions:
nLoss = ATR multiplier x SMA(true range, InpATRLength). The stop ratchets under price while closes remain above it, and above price while closes remain below it. A close through the stop flips it to the other side and draws an arrow.

Translation details:
True range on the first bar uses high-low due to missing previous close. SMA remains undefined until enough bars exist (first value at bar n-1). Previous stop uses “previous stop else current price” while undefined. Persistent state maps to an indicator buffer in MQL5 and an array in Python. Crossovers are strictly false when either side is undefined, affecting initial bars.

Repository: github.com/RAo797bit/pine-python-mql5-conversion

👉 Read | Forum | @mql5dev

Trade Reporter Lite sends MetaTrader 5 trade events to a Telegram chat in real time. It only reports activity and does not open, close, or modify orders. Monitoring covers all positions on the account, including manual trades and any EA. Restarts do not re-announce positions already open.

Open messages include side, symbol, entry, stop, target, pip distances, and lot size. Close messages include pips, R-multiple based on the first stop, P/L in account currency, entry-to-exit, and time held. Failed deliveries retry with back-off at 5 s, 15 s, 45 s, and 135 s. An on-chart status box shows sent, queued, and failed counts plus a readable error.

Setup requires a Telegram bot token from @BotFather and a chat ID (channel @name or numeric ID). WebRequest must allow https://api.telegram.org. In Strategy Tester, messages are written to the journal as TRL[tester].

👉 Read | Quotes | @mql5dev

CME Gap Tracker for gold and Bitcoin measures weekend gaps using CME Friday close and Sunday open, mapped into broker server time with Chicago DST handling. It plots each gap as a box, keeps unfilled gaps extended until price trades back to the Friday close, then marks the fill and records hours-to-fill. Gaps below a configurable minimum percent are ignored.

A panel summarizes recent history: counts by direction, fill rates within 24 hours and 7 days, median and slow-tail fill times, and “heat before fill” in gap-multiples. A built-in fade test enters at the CME open toward the fill, uses a stop sized by gap multiple, times out after N days, charges entry-bar spread, and treats stop/target conflicts conservatively.

Gold backtests on broker spot data showed high fill rates (roughly 83%–97%), but negative results for a fixed fade strategy due to freque...

👉 Read | AppStore | @mql5dev

Gold breakout EA for XAUUSD on H4 uses a Turtle-style entry: buy when the last closed H4 bar finishes above the prior 20-bar high, only if price is above the 200 EMA. Default is long-only, one position at a time, with a 2x ATR(20) stop and a fixed 2R take-profit. Orders are opened at market, then SL/TP are attached to the fill; failure to attach triggers an immediate close. Risk is position-sized from stop distance, spread is filtered (skip when spread >10% of stop), and entries are blocked near session close.

Backtest (Jan 2020–Sep 2026, 10k deposit, 1% risk, spread+swap, random delay) reported +9,029.72 net, PF 1.77, max DD 11.38%, 198 trades, 46.97% win rate. Returns were concentrated in 2024–2025; 2020–2023 ranged from -2.9% to +5.1% per year. Enabling shorts reduced performance in this sample, with sells negative while buys remained positive.

👉 Read | NeuroBook | @mql5dev

Breakout Exit Lab scores Donchian channel breakouts with three exits applied in parallel: ATR trailing stop, Turtle 10-bar exit channel, and a fixed 2R target. Signals are non-repainting and exposed via buffers for EA use, with a panel reporting results in R with spread deducted.

Entry is the first close beyond the prior 20-bar high/low (Turtle System 1), with trades opened on the next bar. Initial stop is 2.0 x ATR(20). Optional 200 EMA trend filter is available. Exit logic includes ATR trail that never loosens, channel exit using prior lows/highs, and fixed target versus the initial stop. Gaps fill at bar open; stop is assumed first when stop and target fall inside one bar.

Tests on XAUUSD showed exit B leading on H4 in 2025 (+66.20R) and 2026 (+22.35R), while H1 leadership varied by year. A 2026 check on EURUSD, GBPUSD, USDJPY was flat to negative...

👉 Read | Freelance | @mql5dev

Renderer changes after Part 4 add a fill-rule switch to the rasterizer, enabling reliable holes and fill-based borders without rewriting the pipeline.

The scanline sweep already computes a signed winding sum from edge crossings. Non-zero treats any non-zero winding as inside. Even-odd treats an odd number of crossings as inside, ignoring direction, so inset contours produce holes without reversing vertex order.

Implementation impact stays minimal: a stable enum with explicit values, a defaulted Fill(rule) parameter to preserve old call sites, and one inside-test change. Even-odd is implemented by testing winding & 1; parity remains correct even with signed increments in two’s complement.

This fixes common UI shapes: glyph counters, donut gauges, ring borders, and stroke-like shapes built as filled contours, while avoiding double-composited anti-alia...

👉 Read | CodeBase | @mql5dev

Most MQL5 position sizers ask for a risk number that is independent of strategy quality. Fixed percent rules size the same whether win rate and payoff are favorable or not.

Kelly sizing derives the risk fraction from the strategy edge: win probability and average win versus loss. A CKelly class can measure these from closed deal history (net of costs), refuse to act on small samples, and return zero sizing when no edge is detected.

Monte Carlo sweeps show the key trade-off: growth peaks at full Kelly, while drawdown and large-loss frequency keep rising. Fractional Kelly, often half or quarter, retains most growth while materially reducing drawdown and ruin risk.

The code is native MQL5: one reusable class plus a sweep script that outputs CSV. Limits remain: edge is estimated, non-stationary, and correlation or overlapping positions are not covered.

👉 Read | AppStore | @mql5dev

“Historical volatility” hides a real implementation choice: close-only, range-based, or full OHLC estimators embed different assumptions. This study benchmarks five variance estimators (Close-to-Close, Parkinson, Garman-Klass, Rogers-Satchell, Yang-Zhang) under a strict persistence forecast: today’s rolling variance becomes tomorrow’s forecast.

Targets are next-session realized-variance proxies built from chronological M1 closes plus the close-to-open jump, with hard rules for session boundaries, gap limits, and endpoint checks. A common target mask ensures every estimator is scored on the same 1,472 EURUSD sessions (2020–2025), avoiding sample drift.

The MQL5 toolkit emphasizes reproducibility: shared OHLC transforms, formula validation, strict chronology safeguards, and paired moving-block bootstrap to test whether observed QLIKE differences (notab...

👉 Read | NeuroBook | @mql5dev

This article replaces clock-based bars with intrinsic time: the market “ticks” only when price reverses by a chosen threshold. An online MQL5 directional-change operator tracks extremes, confirms turns, and splits each move into a directional-change leg plus an overshoot, building a measurable “coastline” from raw ticks.

Using 17.8M EUR/USD ticks, a multi-threshold sweep reproduces key power-law relationships and shows the mean overshoot is roughly the threshold at fine resolutions. A volatility-matched random walk produces nearly identical exponents, suggesting these laws are robust but not a reliable market-structure detector.

The Alpha Engine is then implemented as an MT5 Expert Advisor: a contrarian cascade/de-cascade scheme that adds in fixed steps at intrinsic events, trims on reversals, and controls trend risk with asymmetric thresholds and ...

👉 Read | AppStore | @mql5dev

A non-repainting swing high/low indicator for chart structure and backtesting.

Swings are confirmed only after the required bars to the right have closed. The current forming bar is excluded, so confirmed marks remain unchanged on future candles.

Logic: a swing high is a bar whose high is above the highs of N bars on the left and not below the highs of N bars on the right. Swing lows apply the mirrored rule. This design adds an unavoidable lag of N bars, which is the trade-off for non-repainting behavior.

Key inputs: InpStrength (N, default 5), InpMaxBars (scan depth, default 2000, 0 = all), optional popup alert and mobile push on confirmation.

EA integration: Buffer 0 returns swing high price, Buffer 1 returns swing low price, with EMPTY_VALUE otherwise. Suitable as a neutral building block for BOS/CHoCH logic, not a trading signal.

👉 Read | Forum | @mql5dev

Part 9 extends the MT5 MicroStructure Foundation to fix a key weakness in EMA-stack micro-trend scoring: “trend-aligned” bars can be either fresh breakouts or late retracements, yet both look equally strong.

The solution adds a strategy-neutral pullback depth metric using rolling Fibonacci retracement over the last 20 bars, plus a PULLBACK_QUALITY enum and PullbackAnalysis struct. Depth is filtered by ATR-normalized EMA(5/13) alignment to avoid scoring when no trend exists, and it flags structure as strong/healthy/warning/deep/broken based on how much of the prior swing has been given back.

Two context layers improve signal quality: a 60-bar rolling high/low proxy for H1 range position, and lag-1 return autocorrelation to detect brief momentum persistence. An empirical run on 514 NQ M1 NY sessions finds most bars are shallow pullbacks, while autocorr...

👉 Read | AppStore | @mql5dev

Price Action Volumetric Order Blocks MT5 is an open-source indicator for market structure and order block visualisation in MetaTrader 5. It marks confirmed swing pivots, BOS/MSB events, and tracks active bullish and bearish order blocks projected forward on the chart.

The tool estimates buy/sell participation inside each active zone using bar data and tick volume where needed. These figures are analytical estimates rather than exchange-grade order-flow and should be treated accordingly.

Signals are confirmation-based: pivots, structure breaks, and blocks are only shown after the configured confirmation bars close, reducing repaint-style noise from in-progress candles. A compact panel summarises active zones with price range, buy/sell percentages, and volume.

Key inputs cover pivot sensitivity, maximum active zones, projection length, colour palette,...

👉 Read | Quotes | @mql5dev

Most EA templates still assume one-symbol decision loops. A multi-symbol basket requires explicit data alignment, shared-bar processing, and leg-level execution control.

A basket EA can be built by standardizing each symbol’s close series, running PCA via MQL5 matrix/SVD, and selecting the lowest-variance component as a weight vector. The synthetic spread is the dot product of standardized prices and weights, with deviation expressed as a rolling z-score.

Execution must map basket direction to each leg using weight sign, size each order by abs(weight), and enforce per-symbol min/max/step rules. Orders are sequential, with compensating closes on failure; retcodes must be checked, not only CTrade booleans.

PCA alone does not validate mean reversion, cointegration, or tradability; separate testing is required.

👉 Read | AlgoBook | @mql5dev

NeonDesk is a manual trading panel for MT5. It does not execute automated entries; orders are sent only via BUY/SELL, and each action requires confirmation before submission.

The panel provides a multi-timeframe bias view (M1, M30, H4, D1), a candle countdown timer selectable independently from the chart timeframe (M1 to D1), current Bid/Ask and candle OHLC, plus an ATR-based SL/TP suggestion for reference only.

Trade controls include lot size and SL/TP as direct price levels in the chart’s format, plus a trailing stop in pips that only tightens risk. BUY, SELL, and CLOSE ALL are limited to the panel’s own trades via a dedicated magic number, isolating other positions.

Safety checks include a one-time risk acknowledgement, pre-trade confirmation showing full order details, validation of SL/TP side and broker minimum stop distance, and warnings on existing ...

👉 Read | VPS | @mql5dev

Price charts give quick visual context, but an EA only sees OHLC series. Market context must be derived from swing sequences, not isolated pivots.

A Market Behavior Analyzer can convert detected swing highs/lows into structured objects with metadata: HH/LH/HL/LL label, move length, bar count, retracement, and links to adjacent swings.

Processing is split into stages: swing detection and enrichment, story evaluation, then rendering. The evaluation focuses on the latest confirmed structure, classifies moves as impulse vs pullback relative to that structure, and marks structure as Intact or Threatened using a retracement threshold.

Visualization stays passive: it reads the evaluated state and displays a compact dashboard plus interactive inspection, without re-running analysis.

👉 Read | Calendar | @mql5dev

PropGuard 1.02 is an account-level risk manager for prop-firm accounts. A single instance on any chart monitors all positions on the account, regardless of whether they were opened manually or by other EAs, and enforces daily and maximum loss limits by closing positions and deleting pending orders.

Version 1.02 adds money-based exposure caps, an enforced stop-loss requirement, and a per-position risk cap. In testing against real trade logs, two recurring gaps were highlighted: trades left without a stop loss, and single trades consuming most of the daily loss budget. The new rules can close no-SL positions after a grace period and can tighten stops to a fixed monetary risk per trade.

Maximum loss can be measured as static (initial balance), trailing equity (high watermark), or trailing daily balance (floor updated only at daily reset). Exposure limits are e...

👉 Read | AlgoBook | @mql5dev

ST-Expert reframes market forecasting as a regime-adaptive problem, addressing how correlation structures collapse during shocks like COVID-era shifts and policy cycles. Instead of fitting one stable dependency graph, it maintains multiple specialized “experts” that each represent a distinct market behavior.

Regimes are extracted by splitting history into time intervals that maximize structural dissimilarity using Kendall’s tau, solved via dynamic programming (MSGD). Each interval trains an expert graphon: a probabilistic graph generator that models links between assets as connection probabilities, sampled with Gumbel-Softmax to reduce noisy edges.

Training uses episodic learning: one expert forecasts while a gating network learns to mix the remaining experts to reproduce the active regime. At inference, the gate weights experts from live signals,...

👉 Read | Calendar | @mql5dev