Green Profit gold ea · MT5 XAUUSD… AI Review | QuantTester
Green Profit gold ea · MT5 X: return 48.1%, DD 17.07%, PF 1.24, 352 trades. Read the review.
Strategy Safety Grade: A (85 / 100) · Robust & Controlled
Trading risk analysis result
Strategy Type: Single-Order Trend. Classified as a single-order, non-martingale swing/directional strategy: 352 closed XAUUSD M5 trades over 644 days show a 1.24 profit factor and 17.07% maximum drawdown, but 41.7% of gross profit comes from 18 trades, so expectancy is outlier-dependent and live viability hinges on preserving the long-tail profit engine under real execution costs.
Darwinex 集中度压力测试 (Top 5% Windfall Dependence Test)
剔除排名前 18 笔（5%）极端盈利单后：毛利占比 41.7%，剩余净利润 $-5,503.38，剩余 PF 0.72。
Data transparency & Environment
- MT4/MT5: MT5
- Symbol: XAUUSD
- Timeframe: M5
- Net profit: $4,811.52
- Return: 48.1%
- Max drawdown: 17.07%
- Total Trades: 352
- Profit Factor (PF): 1.24
Critical Algorithmic Vulnerabilities
- Outlier Dependency (Return Consistency Defect): the top 18 trades (5% of 352) generate 41.73% of gross profit; removing them leaves residual net profit of -$5,503.38 and residual PF 0.72, so expectancy is carried by a small set of large winners rather than a uniform per-trade edge.
- Momentum Decay / Protracted Drawdown: equity peaked near $17,719.97 on 2026-06-15 and slid to roughly $14,811.52 by 2026-10-09 (about -16.4%) without reclaiming the high, a multi-month underwater phase that weakens the quote-shape/recovery profile.
- Loss Scale Escalation (Risk Stability): early losses cluster around -$90 to -$115, while from mid-2026 losses repeatedly print -$140 to -$176 (e.g., -176.58 on 2026-06-16, -170.39 on 2026-07-24, -167.04 on 2026-07-06), indicating widened absolute risk per losing trade in the later phase.
- Payoff Asymmetry with Moderate Win Rate: 46.59% win rate means expectancy depends on average winners sufficiently exceeding average losers; in range-bound gold regimes where the large trend expansion fails, the arithmetic edge can flip negative, as the 0.72 residual PF demonstrates.
- Execution Friction Sensitivity: this is an M5 gold system whose average profit per trade is modest (approximately $13.7 over 352 trades); without verified live spread/slippage data, the thin per-trade margin cannot be assumed to absorb real-world cost widening.
Score Breakdown Details
- 【Baseline】Starts at 50 pts neutral baseline
- +15 pts 【Position Discipline】Single fixed/proportional position, no counter-trend averaging
- +5 pts 【Expectancy】Profit factor PF 1.24 with positive expectancy
- +5 pts 【Controlled Drawdown】Max drawdown 17.1% (10%–20%)
- +10 pts 【Superior Calmar】Return-to-drawdown ratio 2.8 (>= 2.5), high edge per unit risk
- Final Strategy Safety Grade: A
Recommended Next Actions
- Export paired entry/exit tickets with entry timestamps and magic numbers to verify whether any contemporaneous basket or scale-in behavior exists.
- Re-run the backtest with realistic commission, swap, and widened spread to quantify how the residual PF 0.72 changes under live friction.
- Request a Demo/Real forward account statement over at least 3–6 months to compare PF, drawdown, and average trade against this backtest baseline.
- Isolate and characterize the 18 top-5% outlier trades (regime, duration, volatility) to determine whether the edge is repeatable trend capture or episodic event dependence.
- Document the stop-loss and trailing-stop configuration and confirm whether the later -$140 to -$176 loss cluster reflects larger stop distances or larger lot sizes.
Q&A About This Report (Quant Due Diligence)
The strategy achieved a Grade A Safety Grade; what are the core strengths and rationale supporting this rating?
The Grade A rating is backed by robust risk controls in the trade data: strict execution discipline (e.g. pre-set stop losses, synchronized basket entries without toxic martingale averaging), and historical drawdowns kept within institutional thresholds.
The equity curve peaked near $17,719.97 in June 2026 and fell about 16% by October 2026 — is that the reported maximum drawdown?
The reported maximum equity drawdown is 17.07%, and the June–October slide from approximately $17,719.97 to $14,811.52 (about -16.4%) is consistent with that order of magnitude. The curve did not reclaim its high by the end of the sample, so the drawdown episode is protracted; the report presents one maximum-drawdown statistic, and the parsed balance path does not establish multiple distinct drawdown episodes beyond this observed decline.
Are the many losing trades in the -$140 to -$176 range evidence of stop-loss failure or unprotected holding?
No. The parsed rows are close-only and do not label exit type, but the losses terminate at bounded values (e.g., -176.58, -170.39, -167.04) rather than compounding without limit, which is consistent with controlled stop exits. The later concentration of larger losses relative to the earlier -$90 to -$115 cluster does suggest widened stop distances or larger lot sizes in that phase, which is disclosed as a risk-stability observation, not as evidence of toxic unprotected holding.
Is the variable lot sizing (0.01 to 0.18) martingale behavior?
The parsed rows are sequential single-out tickets with no verified simultaneous adverse entries, no paired-ticket evidence of averaging down, and no geometric loss-doubling sequence. The lot variation is therefore treated as position sizing or independent signal sizing, not martingale; the four mandatory martingale conditions (same symbol, active floating loss, progressively worse entries, geometric multiplier) cannot be confirmed from close-only evidence.
How is QuantTester different from asking a general-purpose AI assistant myself?
QuantTester parses MT4/MT5 reports into structured metrics and trade records before AI review. Supported calculations can be checked against the original export, guaranteeing consistent audit facts.
Which broker is recommended for trading and running MT4/MT5 strategies?
For algorithmic trading, EA execution and reliable backtesting, we strongly recommend IC Markets (Raw Spread Account) with 0.0 pip spreads.