MQL5 Algo Trading
رفتن به کانال در Telegram
The best publications of the largest community of algotraders. Subscribe to stay up-to-date with modern technologies and trading programs development.
نمایش بیشتر📈 تحلیل کانال تلگرام MQL5 Algo Trading
کانال MQL5 Algo Trading (@mql5dev) در بخش زبانی انگلیسی بازیگری فعال است. در حال حاضر جامعه شامل 520 912 مشترک است و جایگاه 149 را در دسته فناوری و برنامهها و رتبه 5 را در منطقه المملكة المتحدة دارد.
📊 شاخصهای مخاطب و پویایی
از زمان ایجاد در невідомо، پروژه رشد سریعی داشته و 520 912 مشترک جذب کرده است.
بر اساس آخرین دادهها در تاریخ 13 ژوئیه, 2026، کانال فعالیت پایداری دارد. در ۳۰ روز گذشته تغییر اعضا برابر 11 174 و در ۲۴ ساعت گذشته برابر 692 بوده و همچنان دسترسی گستردهای حفظ شده است.
- وضعیت تأیید: تأیید نشده
- نرخ تعامل (ER): میانگین تعامل مخاطب 3.36% است و در ۲۴ ساعت نخست پس از انتشار، محتوا معمولاً 1.87% واکنش نسبت به کل مشترکان کسب میکند.
- دسترسی پستها: هر پست به طور میانگین 17 508 بازدید دریافت میکند. در اولین روز معمولاً 9 729 بازدید جمعآوری میشود.
- واکنشها و تعامل: مخاطبان بهطور فعال حمایت میکنند؛ میانگین واکنش به هر پست 38 است.
- علایق موضوعی: محتوا بر موضوعات کلیدی مانند indicator, chart, mql5, candle, range تمرکز دارد.
📝 توضیح و سیاست محتوایی
نویسنده این فضا را محل بیان دیدگاههای شخصی توصیف میکند:
“The best publications of the largest community of algotraders.
Subscribe to stay up-to-date with modern technologies and trading programs development.”
به لطف بهروزرسانیهای پرتکرار (آخرین داده در تاریخ 14 ژوئیه, 2026)، کانال همواره بهروز و دارای دسترسی بالاست. تحلیلها نشان میدهد مخاطبان بهطور فعال با محتوا تعامل دارند و آن را به نقطه اثرگذاری مهم در دسته فناوری و برنامهها تبدیل کردهاند.
520 912
مشترکین
+69224 ساعت
+3 7007 روز
+11 17430 روز
در حال بارگیری داده...
کانالهای مشابه
ابر برچسبها
اشارات ورودی و خروجی
---
---
---
---
---
---
جذب مشترکین
ژوئیه '26
ژوئیه '26
+5 856
در 5 کانالها
ژوئن '26
+9 701
در 6 کانالها
Get PRO
مه '26
+10 784
در 10 کانالها
Get PRO
آوریل '26
+10 664
در 10 کانالها
Get PRO
مارس '26
+22 886
در 10 کانالها
Get PRO
فوریه '26
+25 031
در 6 کانالها
Get PRO
ژانویه '26
+26 330
در 10 کانالها
Get PRO
دسامبر '25
+22 353
در 4 کانالها
Get PRO
نوامبر '25
+21 953
در 11 کانالها
Get PRO
اکتبر '25
+24 933
در 4 کانالها
Get PRO
سپتامبر '25
+19 967
در 5 کانالها
Get PRO
اوت '25
+17 145
در 4 کانالها
Get PRO
ژوئیه '25
+18 909
در 10 کانالها
Get PRO
ژوئن '25
+18 135
در 9 کانالها
Get PRO
مه '25
+17 357
در 2 کانالها
Get PRO
آوریل '25
+19 677
در 0 کانالها
Get PRO
مارس '25
+19 081
در 0 کانالها
Get PRO
فوریه '25
+18 865
در 1 کانالها
Get PRO
ژانویه '25
+19 658
در 0 کانالها
Get PRO
دسامبر '24
+17 881
در 0 کانالها
Get PRO
نوامبر '24
+23 274
در 0 کانالها
Get PRO
اکتبر '24
+25 611
در 3 کانالها
Get PRO
سپتامبر '24
+23 810
در 0 کانالها
Get PRO
اوت '24
+32 845
در 0 کانالها
Get PRO
ژوئیه '24
+39 916
در 0 کانالها
Get PRO
ژوئن '24
+14 429
در 0 کانالها
Get PRO
مه '24
+1 105
در 0 کانالها
Get PRO
آوریل '24
+1 407
در 0 کانالها
Get PRO
مارس '24
+2 180
در 0 کانالها
Get PRO
فوریه '24
+1 456
در 0 کانالها
Get PRO
ژانویه '24
+1 276
در 0 کانالها
Get PRO
دسامبر '23
+4 228
در 0 کانالها
| تاریخ | رشد مشترکین | اشارات | کانالها | |
| 14 ژوئیه | 0 | |||
| 13 ژوئیه | +692 | |||
| 12 ژوئیه | +366 | |||
| 11 ژوئیه | +397 | |||
| 10 ژوئیه | +476 | |||
| 09 ژوئیه | +475 | |||
| 08 ژوئیه | +567 | |||
| 07 ژوئیه | +735 | |||
| 06 ژوئیه | +691 | |||
| 05 ژوئیه | +311 | |||
| 04 ژوئیه | +184 | |||
| 03 ژوئیه | +207 | |||
| 02 ژوئیه | +459 | |||
| 01 ژوئیه | +296 |
پستهای کانال
Work continued on MT5 replay/simulation tooling to render position state directly on the chart, not in terminal text. Standard platform position visuals are insufficient under cross-order execution and replay models.
The update adds three HLINE objects per position: open price, stop-loss, and take-profit. Object naming uses the position ticket plus suffixes to keep uniqueness and support bulk cleanup on indicator removal.
A priority enumeration was introduced to reduce selection issues when objects overlap. C_Terminal and C_ChartFloatingRAD were adjusted to consume the shared priority list, keeping Chart Trade above most graphics while allowing order/position overlays to remain selectable.
Further refinements add OBJPROP_TEXT descriptions for each line so the Objects list remains self-explanatory without changing user chart settings.
👉 Read | CodeBase | @mql5dev
| 2 |  MetaTrader 5 object interactivity is extended from simple selection outlines to controlled resizing workflows.
Current implementation highlights a selected object but lacks usable grab handles because helper objects share a protected name prefix, making them non-selectable. Forcing initial selection on specific objects enables MT5 drag behavior, but default anchor handling creates inconsistent resizing visuals and disconnected outline segments.
The update adds dedicated corner points and uses switch fall-through to reuse creation logic while adjusting placement, producing two effective resize handles.
Event handling is then tightened: clicking should move objects without toggling their selected state. A small handler addition blocks selection-state changes while keeping drag operations active.
👉 Read | Forum | @mql5dev | 2 749 |
| 3 |  An open-source EA focuses on Stochastic-based position closing with three modes. Crossing Mode closes in OB/OS when %K and %D cross. Entering Mode closes when %D enters overbought or oversold. Exiting Mode closes when %D exits those zones. A profit-state filter restricts closures to Loss, Profit, or both.
Core parameters cover %K/%D periods, slowing, OB/OS levels, price field, and a selectable Stochastic timeframe independent of the chart. Logic runs on every tick and affects only the current symbol.
Optional Test Mode adds simple entries for research: 0.01-lot trades when %D crosses 50 with a configurable offset, only when no positions exist. No SL/TP, no magic filtering, and it can run alongside the closing logic. No money management features are included.
👉 Read | VPS | @mql5dev | 4 143 |
| 4 |  Indicator buffer data is being formalized as a first-class timeseries: per indicator, maintain a list containing all buffer values for every bar used in calculations. This enables consistent storage, search, sort, and later statistical analysis across library collections.
Core library changes: the NewBar class is relocated to Services; SeriesDE include paths are updated; new message indices/texts are added; a dedicated collection ID and timer constants are introduced. Indicator data gains an extra integer property for indicator handle, including sorting/search criteria support.
New class CSeriesDataInd stores CDataInd objects per bar/buffer, supports creation, refresh on current bar, append on new bar, and trimming to required depth. Indicator objects are extended with total buffer count and an embedded buffer-data series; the indicators collection and eng...
👉 Read | AlgoBook | @mql5dev | 13 087 |
| 5 |  BBandsPsar is a custom hybrid indicator that merges a volatility model (Bollinger Bands) with a trend-following signal (Parabolic SAR) into one output.
The core calculation measures the gap between the Parabolic SAR value and the current candle’s open or close, reflecting SAR’s alternating behavior during trend changes. That gap is then normalized using Bollinger Bands, producing a standardized histogram rather than raw price-distance values.
Because the output is scaled, the same threshold logic can be applied across instruments with different price ranges, improving cross-asset comparability.
BBandsPsar inherits the full input sets of both Bollinger Bands and Parabolic SAR, allowing adjustment of period, deviation, price source, step, and maximum settings to control sensitivity.
For dynamic parameter testing, the iBands call can be refactored so ...
👉 Read | Freelance | @mql5dev | 12 189 |
| 6 |  Backtests only charge the costs that were configured, so net profit and profit factor do not show how close a strategy is to the execution-cost boundary. This MQL5 tool measures the limit directly: how much cost the strategy can absorb before the edge disappears.
It reads a CSV (Date,Profit,Volume; one row per closing deal) and prints a report to the Experts tab. Output includes breakeven cost per deal, cushion versus an assumed realistic cost, net and profit factor re-priced at the assumed cost, a cost-sensitivity curve from 0 to 3x, and win erosion where costs flip winners into losers. A composite A+ to F score aggregates cushion, profit-factor resilience, and erosion, with recommendations.
Input file goes to MQL5\Files; Volume is optional. If missing on first run, a reproducible sample file is generated. A helper ExportCost.mq5 can export deals from ...
👉 Read | CodeBase | @mql5dev | 11 966 |
| 7 |  Spreadsheets can be used as a lightweight research harness for algorithmic strategies: import MT5 history from CSV/TXT, normalize numeric formats, and keep the dataset small enough (around a few thousand rows) to avoid slow recalculation with minimal impact on results.
The workflow builds indicators directly in cells. A separate “variables” sheet stores parameters, while functions like IF, AVERAGE and INDIRECT generate a configurable SMA over dynamic ranges. Relative vs absolute references ($) make formulas safe to copy across long columns.
A moving-average crossover is then modeled as stateful trade logic: one column detects crossings, another persists position status bar-by-bar, and a signal column maps actions (Buy/Sell/Close) with correct next-bar execution. Additional columns propagate entry price (including spread) and compute pip outcomes for late...
👉 Read | AppStore | @mql5dev | 17 021 |
| 8 |  BBandsPsar is a hybrid indicator combining Bollinger Bands volatility metrics with Parabolic SAR trend signaling in a single output.
The calculation tracks the distance between the Parabolic SAR level and the candle open/close, capturing SAR’s alternating behavior during trend changes. This gap is then normalized through Bollinger Bands and rendered as a histogram.
Band-based standardization targets more stable thresholds across instruments, supporting cross-asset comparison on a unified scale. All configuration inputs from both Bollinger Bands and Parabolic SAR are retained, allowing adjustment of sensitivity, smoothing, and switching behavior to match market conditions.
👉 Read | Quotes | @mql5dev | 14 299 |
| 9 |  This article upgrades the classic SuperTrend by making its ATR multiplier responsive to momentum divergence, using MPO4 (or optionally RSI) as the trigger. When a validated bullish/bearish divergence appears, a “shrinking” rule reduces the multiplier by a sensitivity factor, pulling the band closer to price. The result is tighter trailing stops near exhaustion and earlier flips for capturing the next trend.
A key focus is non-repainting behavior in MT5 indicators. Instead of fragile global flags, the design persists divergence context per bar via state buffers (last pivot prices/osc values, last divergence type, bars-since counter). State is propagated forward each iteration and reset on trend flips, keeping historical signals stable across ticks, refreshes, and restarts.
Implementation details cover a chart-window MQL5 indicator with a colored SuperTrend...
👉 Read | Docs | @mql5dev | 15 508 |
| 10 |  Price-window indicators in MQL5 often shift arrays with ArrayCopy() on every bar to keep index 0 as “latest”. That design is O(n) per update and scales poorly with large periods, multi-symbol sets, and dense OnCalculate() workloads.
A circular buffer removes the shift. Only a head index advances with modulo arithmetic, and one slot is overwritten. Push and Get become O(1) regardless of window size; allocation is fixed at construction.
An MQL5 template CCircularBuffer<T> fits this pattern but comes with constraints: no specialization, template code must stay in-header, and capacity must be runtime. Statistical routines should use two-pass variance to avoid cancellation common in price series.
Result: less memory movement per bar, improved backtest throughput, and reduced live tick latency under load.
👉 Read | Signals | @mql5dev | 13 715 |
| 11 |  False breaks of recent N-bar highs/lows often act as liquidity sweeps: price runs stops beyond an obvious extreme, then closes back inside the range. The article turns this into a fully testable Turtle Soup contract: minimum sweep depth (points and % of range), minimum age of the swept level, sweep-window rules, and close-based rejection confirmation (multiple closes, optional reversal body).
The MQL5 implementation focuses on robustness: bar-close execution, duplicate-signal guards, per-direction position caps, and broker stop-level compliance. Stops can be dynamic beyond the sweep extreme or fixed-point; targets support R-multiple projection or the opposite side of the lookback range, with optional trailing.
It also builds chart tooling (levels, sweep zones, risk/reward boxes) and a scan engine using iHighest/iLowest and close checks, ready for S...
👉 Read | CodeBase | @mql5dev | 12 185 |
| 12 |  Mean-reversion implementation in MetaTrader 5: weekly channel breaks using two moving averages on High and Low with a shared period. Break above the upper band triggers short bias; break below the lower band triggers long bias.
Signal confirmation uses an ONNX-backed statistical model plus daily open/close moving averages for directional filtering. Data pipeline exports EURUSD daily bars to CSV via MQL5, then trains in Python on 2011–2019 and tests on 2020–2026. Model selection highlights overfitting in flexible learners; a linear regressor is chosen and exported to ONNX.
The EA loads the ONNX model, manages trades via magic number filtering, updates indicator buffers, and applies ATR-based symmetric SL/TP with trailing stops. Backtests with real ticks and random delay show modest positive expectancy, profit factor near 1.1, and larger average win...
👉 Read | Calendar | @mql5dev | 11 394 |
| 13 |  Most MT5 risk tools implicitly cap worst-case planning to what the recent window already contains. ATR, bootstrapped trade resampling, and historical VaR all operate on observed volatility and outcomes, so tail losses beyond the sample are systematically understated.
Financial returns are fat-tailed, so rare moves occur more often than normal assumptions imply. Extreme Value Theory addresses this by modeling only the tail, allowing extrapolation beyond the historical maximum loss.
An MQL5-native implementation can use Peaks-Over-Threshold with a Generalized Pareto fit via ALGLIB MLE. Outputs include EVT VaR, Expected Shortfall, and a shape parameter that quantifies tail heaviness, with hard refusal to report metrics when exceedances are insufficient.
👉 Read | Signals | @mql5dev | 11 402 |
| 14 |  Protecting profit after a position is opened typically requires rules that react to price movement, not just the entry signal. Common approaches include moving the stop loss to breakeven after a defined profit threshold, then switching to a trailing stop to lock in gains as volatility expands or contracts.
A structured method uses staged stop management: initial fixed stop, breakeven activation, and trailing based on points, ATR, or recent swing levels. Partial closes can reduce risk while keeping exposure for continuation. Controls such as minimum distance, step size, and spread filters help avoid premature stop-outs in noisy conditions.
👉 Read | AlgoBook | @mql5dev | 10 897 |
| 15 |  Replay/simulation work in MQL5 is moving past sockets and SQL basics into chart-side tooling needed for realistic interaction.
Current components (mouse pointer, Chart Trade indicator, EA messaging) can submit market execution, but cross-orders and replay symbols lack reliable visual feedback for positions and orders.
Next step is a minimal position-visualization indicator. An indicator can read position state and print details, without attempting trade actions, which remain EA-only. This also highlights a platform constraint: indicators share a single chart thread, so the logic must avoid blocking and unnecessary updates.
Focus shifts to controlled chart objects for position lines, with attention to object lifecycle and cleanup to prevent long-term chart degradation.
👉 Read | Docs | @mql5dev | 10 379 |
| 16 |  MetaTrader 5 hides powerful “indicator-like” behavior inside chart objects. By programmatically repurposing OBJ_FIBO, the article builds a custom risk tool that visually marks entry, stop, and target using only a few configured levels.
The core technique sets a fixed number of Fibonacci levels, then assigns each level its value, color, and style via parallel arrays. Using values outside 0..1 turns Fibonacci into a projection engine, making it easy to express 1:1, 1:1.5, and partial-exit layouts that can’t be recreated manually from the UI.
Practical refinements include locking selection to prevent accidental moves, hiding unwanted guide lines, exposing a stop/target ratio as an input, and optionally disabling right-side extension so multiple analyses can coexist cleanly. The article also explores switching drawing flow toward left-click interaction, ...
👉 Read | Signals | @mql5dev | 9 798 |
| 17 |  This article extends the MCP approach from trading execution to the full development workflow by connecting AI agents to MQL5 Algo Forge, a Git-based host backed by Forgejo. Using its HTTPS REST API, an assistant can programmatically create repos, commit EA files, manage branches, open pull requests, file issues, and publish releases.
The server is a portable Python project (no MetaTrader/Windows dependencies) built in layers: JSON config + token, an httpx-based API client with consistent error normalization, domain handlers, and a FastMCP tool surface exposing 12 tools.
Key implementation details include safe token handling (config or environment variable) and Base64 encoding/decoding for file endpoints. A single “commit file” tool abstracts POST vs PUT by detecting existing files and retrieving sha automatically, enabling reliable updates without t...
👉 Read | Docs | @mql5dev | 9 286 |
| 18 |  Reliable MT5 Expert Advisors require systematic validation before sending trade requests. Common constraints include max positions/orders, volume min/max/step, SL/TP distance, margin, session limits, symbol permissions, and news filters. Market publication tests already enforce many of these cases, so reusable checks reduce duplication and regressions.
A compact MQL5 validator set can cover: lot sizing with step-based normalization (avoids retcode 10014), SL/TP distance vs SYMBOL_TRADE_STOPS_LEVEL, price digit normalization (avoids retcode 10015), freeze-level checks for order modifications, margin/funds checks, and “no changes” guards to prevent TRADE_RETCODE_NO_CHANGES.
Additional utilities include: pending-order limits, lightweight new-bar detection, tradability checks for SYMBOL_TRADE_MODE_DISABLED, calendar-based news windows (live only), and UTC sess...
👉 Read | Signals | @mql5dev | 9 598 |
| 19 |  This part adds a pinned-tools ribbon to complement the deep, tabbed sidebar: the sidebar stays optimized for discovery, while the ribbon provides one-click access to a small set of frequently used drawing tools. The ribbon auto-hides when empty and preserves pin order.
Pinned tools are stored in the engine as an ordered list with a minimal API (count, get, contains, pin/unpin, toggle). Pinning appends; unpinning compacts the array without reordering, keeping behavior predictable.
A shared anti-aliased pushpin glyph is rendered via polygon coverage and alpha blending, reused consistently across flyout rows, the Pinned sidebar tile, and the ribbon.
The ribbon is a floating surface with drag, resize, and horizontal scrolling. It clips overflowing icons using an offscreen canvas and shows a proportional scrollbar thumb, keeping interaction smooth eve...
👉 Read | NeuroBook | @mql5dev | 10 791 |
| 20 |  Walk-Forward Analysis is reframed as a measurable robustness test for MT5 EAs: optimize on in-sample windows, score degradation on forward windows, and summarize it as a reproducible metric instead of eyeballing equity curves.
The core score is Walk-Forward Efficiency (WFE): per-window ratio of out-of-sample to in-sample Sharpe. Windows pass only if they retain at least 50% efficiency, with guards that force failure when in-sample Sharpe is non-positive or too small to be meaningful.
A native MQL5 pipeline implements the full loop: an EA logs per-bar equity to CSV, a fast reader ingests it, WFE_Engine.mqh computes Sharpe/WFE with numerical stability and validity flags, and a CCanvas histogram renders pass/fail windows and reference lines directly on-chart for immediate diagnosis.
👉 Read | Calendar | @mql5dev | 10 914 |
