Third-party product
MT4 Gold EA on XAUUSD M5 - Independently Backtested, Results Published in Full
TL;DR
MT4 XAUUSD M5 EA (Quantum Emperor v5.8). Verified Strategy Tester run on an IC Markets demo: 610 trades, 95.41% win rate, but profit factor only 1.39 and maximum drawdown 58.48%. Modelling quality 64.17% with 25 mismatched-chart errors, so the simulation itself is low fidelity. Results are not a promise of future performance.
Independently tested
RBOT.CC provides the review conditions and public evidence report below. Results are not a promise of future performance.
This page describes a third-party product. RBOT.CC is not the developer unless explicitly stated. Verify all claims before use.
QuantTester safety grade A 85/100 · 75-89 pts
Strategy Type: Hybrid (Tiered Basket + Trend Following). Classified as Layered Basket (Tiered Basket execution): a 863-day M5 XAUUSD backtest of Quantum Emperor V5.8 shows a 96.6% win rate and 1.04 profit factor on disciplined six-to-seven-unit basket exits, but the +1.02% net return and 14.98% maximum drawdown rest on an extreme payoff asymmetry (0.04) and demonstrable outlier dependency, leaving the strategy viable only after forward execution-cost validation.
Performance figures depend on the test period, broker conditions, spread, commission, slippage, leverage, and risk settings. Confirm the test dates and source report before relying on any figure. Backtest results are not a promise of future performance.
Quantum Emperor MT4 (self-identified in this report as "Quantum Emperor MT4 v5.8, 13 Oct 2024") is a box-formation EA: during a daily trading box (03:00-11:00 in this run) it places six pending stop orders on each side (Amount 6) around the price, cancels the opposite side once one side fills (cancelopposite), expires unfilled pendings at 21:30, and closes profitable positions at a fixed time (10:59) via CloseTradesInProfit. It carries a SmartRecovery module that multiplies size by 1.3, up to 3 times, after losing sequences - a recovery/martingale-style feature that increases exposure precisely when the account is under stress. Base lot is 0.01 with risklevel 7 and StopLoss 2500 (points). Notably, the EA's own description input reads "Expert Advisor for GBPUSD", while this test - like the vendor's marketing - ran it on gold. The Strategy Tester run documented here: ICMarketsSC-Demo03 demo server (build 1492), XAUUSD on M5, every-tick model, configured window 2026.01.03 - 2026.10.12, modelled 2026.01.05 01:00 - 2026.10.09 23:55, 52,561 bars, 53,520,605 ticks modelled, fixed spread 10, initial deposit 10,000.00. Important quality flags: modelling quality is only 64.17% and the report lists 25 mismatched-charts errors, so this simulation is materially lower fidelity than the vendor's own published backtest (90.06%). Closed trades span 2026-01-05 to 2026-07-02 only - 610 trades in roughly six months. From 2026-07-02 until the end of the test the EA kept placing its pending orders, but every one was deleted unfilled, so roughly the final three months of the window contain no realized trades. Within the active period it won 95.41% of trades (582 wins / 28 losses; longs 93.67%, shorts 97.48%) with a profit factor of just 1.39 and a net profit of 12,886.96 (+128.87%).
The headline win rate is real but almost meaningless on its own. The average win is 79.05 while the average loss is -1,182.86 - a payoff ratio of about 0.07, meaning each losing trade erases roughly fifteen winning trades. The largest single loss was -1,805.04 versus a largest win of 377.85. This is the classic profile of a box/grid EA with stop-loss clusters: most exits are small gains harvested from a stop line that has been moved into profit (560 of 610 exits carry an s/l comment, 50 a t/p comment), while the rare full stop-out hits six layered orders at once. Those clusters are visible in the data: on 2026-02-03, 2026-02-19, 2026-05-26 and 2026-06-17, six same-minute stop-loss exits landed together (for example -1,529.27 x6 plus -1,236.56 on 2026-06-17). The maximum drawdown was 20,251.90, or 58.48% of the account - the balance fell from a peak near 34,318 to about 18,858 around the June cluster and closed the sample at 22,886.96. Three of the seven active months were negative (Feb -4,916, May -3,711, Jun -2,697). The maximum winning streak was 305 trades, which inflates the perceived safety of a system whose worst week can remove most of a year's gains. Three further caveats. First, data fidelity: 64.17% modelling quality and 25 mismatched-charts errors mean the win rate, drawdown and return can all shift materially in a clean-data rerun. Second, costs: the test used a fixed spread of 10 while the vendor has recommended a 2.0 spread; wider news-time spreads on gold would hit this strategy harder than the simulation shows, and commission and swap are not stated in the report. Third, the vendor figures previously published on this page - 96.96% win rate, 2.44 profit factor, 12.35% drawdown over 2,726 trades - belong to a different, larger backtest (the images previously used here show 2,918 trades, 972.24% net profit, 24.59% balance drawdown, 90.06% modelling quality). They do not describe this run and have been removed. No demo or live forward record was supplied; nothing here is a promise of future performance.
Exact inputs used in the test above. Default EA settings will not reproduce these results — import these values before running it live.
| Parameter | Value |
|---|---|
| Auto | false |
| risklevel | 7 |
| Lot | 0.01 |
| MagicNumber | 777 |
| SmartRecovery | true |
| SmartRecoveryMultiplier | 1.3 |
| SmartRecoveryMultiplierTimes | 3 |
| additionaltrade | true |
| CloseTradesInProfit | true |
| tradeFriday | 1 |
| DisableNFP | true |
| DisableTradeHoliday | true |
| Slippage | 5 |
| EnableDST | true |
| AbPinp | 1 |
| BePinp | 1 |
| StopLoss | 2500 |
| Amount | 6 |
| cancelopposite | true |
| CancelTime | 0 |
| Bars in test | 84086 |
| Mismatched charts errors | 56 |
.ex4 for MT4 or .ex5 for MT5).set) for supported currency pairsquick_install.bat)quick_install.bat.MQL4/Experts (or MQL5/Experts) and preset files into MQL4/Presets.Yes. All core indicators and EAs on RBOT.CC are free to download directly without email walls, hidden paywalls, or account limits. If a third-party developer offers an optional commercial Pro version, it is clearly indicated with an external source link.
Yes. You can run them on unlimited demo and live accounts with any MetaTrader broker. We strongly recommend testing thoroughly on a demo account for at least 2 to 4 weeks to observe trade behavior and drawdowns before using real money.
You can join our Telegram community (@wwwrbotcc) to ask setup questions, share backtests, and discuss strategy settings with the RBOT team and other algorithmic traders.
⚠ READ BEFORE USE
This EA is for educational and expert use only. Test on a DEMO account for at least 30 days before going live. By clicking “OK”, you acknowledge and accept the risks of trading.