🥇 GOLD EA (XAUUSD M30 · REGIME-FILTERED BREAKOUT)
GOLD NEURON EA
Three engines work together in one account: a regime-filtered trend core and a momentum breakout cover each other's weak market conditions, while an ML pivot entry gate (default ON since v3.13) filters entries. Signals are computed on a fixed M30 timeframe and are independent of your chart timeframe. Equal-lot grid up to 3 positions, no martingale, no averaging down. No external files or DLLs; runs fully offline.
🔄 v3 replaced the engine entirely — older figures are withdrawn
With v3 this product moved from a single AI-decision design to a multi-engine architecture, and v3.13 adds an ML pivot entry gate as a third engine. The figures published for the v1 line (once-daily decisions, the high Y2025 multiples) are withdrawn. Every number on this page comes from a 22.6-year (2004-2026) 100% real-tick measurement of the two core engines (trend core + momentum breakout).
⚠ Aggressive EA (high risk) — please read
The shipping default (Ultra) reached a 38.2% equity drawdown over the 22.6-year measurement. If you want lower risk, choose Standard (29.5% equity DD) or Defensive. Trade only money you can afford to lose, and always start on demo or with a small account.
⚡ Minimum deposit: from $100
XAUUSD, any chart timeframe (signals are computed on a fixed M30), 1:500 leverage recommended. Compounding percent-of-equity sizing follows your balance automatically. For small or conservative accounts, use a micro (cent) account with Defensive mode.
Profit Factor
1.28
期待値
勝率
41.4%
1,629取引
最大DD
38.20%
証拠金DD
年率複利
+17.1%
+$1,710/年（$10,000基準）
Attach to: XAUUSD M30 — The strategy reads the M30 timeframe. On any other chart it will place no trades at all. When testing in the MT5 Strategy Tester, set the symbol and timeframe to XAUUSD M30 as well.
HOW IT WORKS — REGIME-FILTERED BREAKOUT
How it works — three engines in one account
Engine A (trend core) uses a regime filter to trade pullbacks only in trending conditions; Engine V (VIPER) trades momentum breakouts out of volatility contraction. Because they favour different market states, running both in one account spreads the opportunity set.
- Entry: each engine decides independently (signals computed on M30 bars, chart-TF independent). Same-direction additions are capped at an equal-lot grid of 3.
- Exit: TP/SL bracket on every position plus trailing. No lot escalation, no averaging down.
- Risk: compounding percent-of-equity sizing with margin and volume-cap guards. The risk budget is split equally across enabled engines, and the RunMode dropdown scales the whole thing across four levels.
- No external files, DLLs or API calls — runs offline. An economic-calendar filter (NFP/FOMC avoidance) is built in.
ONE DROPDOWN — FOUR RISK PRESETS
RunMode dropdown — four risk presets
| Preset | Risk × | Return (real tick, 2025 full year) | Max equity DD |
|---|---|---|---|
| Defensive | 0.5 | — | — |
| Standard | 1.0 | x24.1 | 29.5% |
| Aggressive | 2.0 | — | — |
| Ultra (default) | 2.0 | x35.2 | 38.2% |
All presets share the same signals and differ only in risk multiplier. Figures are measured on XAUUSDduka, 100% real ticks (Model=4), 2004.01.01-2026.07.31 (22.6 years), $10,000 start, 1:500, default inputs. Defensive and Aggressive are shown as — because no report over the same window has been produced for them yet; estimates are not published here. Ultra is the measured ceiling for the risk multiplier — going beyond it collapses in over-Kelly territory.
VALIDATION — 22.6 YEARS, 100% REAL TICKS
Validation — 22.6 years, 100% real ticks
Measured end-to-end on Dukascopy-sourced real ticks (XAUUSDduka) with the MT5 Strategy Tester in Model=4 (100% real ticks) across 2004.01.01-2026.07.31, 22.6 years, using default inputs, a $10,000 start and 1:500. Cross-validation on a third-party feed and Monte Carlo testing have not been run for the published two-core-engine configuration, so they are not shown here.
| Preset | Window | Total return |
|---|---|---|
| Ultra (shipping default) | 22.6y | +3,421% |
| Standard | 22.6y | +2,312% |
Note: because signals are computed on a fixed M30 timeframe, differences in chart timeframe or broker GMT offset do not affect them.
All figures above are backtest measurements and do not guarantee future performance.
Verification Status
How far verification has progressed, from backtest to forward testing
- BacktestDone
- Trades
- 1629
- PF
- 1.28
- Max DD
- 38.20%
- OOSDone
- Data source
- MT5 official ticks
- Years verified
- 22.6
- ForwardIn progress
- Broker
- EXNESS
- Started
- Jul 17, 2026
- Trades
- 35
- PF
- 1.02
- Max DD
- 39.38%
- Gain
- +1.46%
- Updated
- 5 hours ago
- LiveNot started
📈MT5 Strategy Tester Report — GOLD NEURON EA
📋 About this report
- Official report generated by the MT5 Strategy Tester (real-tick data)
- Multi-year real price data with the every-tick (real ticks) model
- Balance curve (blue) / Equity curve (green) shows floating losses too
- Past results do not guarantee future profits
3-ENGINE BACKTEST — SUPPLEMENTARY DISCLOSURE
3-Engine Backtest (Supplementary Disclosure)
Supplementary backtest with all three engines enabled (Trend + Momentum BO + ML Pivot, RunMode=Ultra). Conditions: XM (symbol GOLD) / M30 / Model=1 (1-minute OHLC — NOT real ticks) / 2022.01–2026.07 (4.5 years) / initial deposit 10,000 USD / includes the ML in-sample training period. No htm report was produced; figures were reconstructed from the Strategy Tester log trade stream (only the final balance is the official value).
Results: final balance 15,276.39 USD (+52.8% / 4.5 years, simple annual ≈11.7%), 477 trades, PF 1.22 (price-diff basis; lower incl. swap), win rate 59.5%, max balance drawdown 35.3%.
Not directly comparable to the published figures on this page (100% real ticks, 22.6 years, two core engines). The published figures remain unchanged.
View the full reconstructed report (HTML)⚠ Where it struggles — main risks
- The two core engines trade direction and breakouts, so extended low-volatility chop with no clear direction accumulates small losses.
- The shipping default Ultra reached 38.2% equity DD over 22.6 years — roughly 40% of the account in open loss. Choose Standard or Defensive if that is unacceptable.
- Ultra is the measured ceiling for the risk multiplier; raising it further has been measured to break down in over-Kelly territory. Modifying inputs upward risks ruin.
- Backtests are past results and do not guarantee future profits. Live track record is still accumulating.
Download (free)
STEP 1 Prepare your account
Using the EA requires an MT5 account opened through our referral link. If you already have an account, please create an additional account.
No account yet — open a new account via the links below
Adding an account — create an additional account in your member pageSee how →
Buy GOLD NEURON EA
Buy on the official MetaQuotes Market (MQL5 Market). Payment, licensing and automatic updates are all handled inside the MT5 app.
🏷 Launch pricing now — the price steps up every 10 copies sold. The earlier you buy, the lower the price.
Buy on MQL5 Market →Buy on the official MetaQuotes Market — install directly in MT5 with automatic updates. No account registration needed.
Figures shown are backtest measurements on 100% real ticks (Model=4) — MT5 Strategy Tester, default inputs, XAUUSDduka, 2004-2026. They do not guarantee future results. Automated FX trading carries the risk of losing your capital; use risk capital only.