Quantum Emperor V7.5 EA · MT5 XAUUSD H1… AI 测评｜QuantTester
Quantum Emperor V7.5 EA · MT5 XAUUS 报告：收益 -0.8%、回撤 3.43%、PF 0.94、303 笔交易。AI 风险解读。
策略安全信用评级: E (0 / 100) · 高危预警 (High Risk)
交易风险分析结果
Strategy Type: Grid / Recovery. Classified as Grid / Recovery (adverse grid scale‑in with 'Smart Recovery'): the account closed 303 trades for a net loss of −83.04 (−0.83%) on a 10,000 base, with a 0.94 profit factor, an average win of 4.76 versus an average loss of −27.02 (payoff ratio 0.18), and residual PF of 0.83 after removing the top 5% of trades, demonstrating that a high 84.16% win rate is entirely absorbed by a small number of large stop‑outs and that expectancy is negative and structurally fragile.
Darwinex 集中度压力测试 (Top 5% Windfall Dependence Test)
剔除排名前 16 笔（5%）极端盈利单后：毛利占比 11.05%，剩余净利润 $-217.24，剩余 PF 0.83。
数据透明度与环境说明
- MT4/MT5: MT5
- 品种: XAUUSD
- 周期: H1
- 总净利润: $-83.04
- 总收益率: -0.8%
- 最大回撤: 3.43%
- 总交易笔数: 303
- 盈利因子 (PF): 0.94
AI 识别的核心高危漏洞清单
- 【Return Consistency Defect / R+】Outlier Dependency: The top 5% of trades (16 rows) account for 11.05% of gross profit; removing them leaves a residual net profit of −217.24 and a PF of 0.83, indicating the strategy's edge relies on a small set of outliers rather than a systematic expectancy.
- 【Momentum Decay / D‑Score Quote Shape Broken】Protracted Underwater Drift: The balance curve declines from a peak near 10,184 (achieved by late 2025‑04) to 9,916.96 by 2026‑10‑08, with no sustained recovery above the prior high; this is a persistent negative momentum regime consistent with alpha decay.
- 【Risk Stability Breakdown / Rs】Loss‑Scale Clustering and Recovery Sizing: Later losses cluster around −25 to −41 (e.g. −41.40 on 2026‑02‑17, −36.81 on 2026‑02‑04, −33.48 on 2026‑03‑23), while average win is only 4.76; the 'Smart Recovery 1.6 3' setting suggests loss‑driven sizing expansion under stress, undermining risk stability (Rs).
- 【Payoff Asymmetry Fragility / La】Win‑Rate Dependence: An 84.16% win rate with a 0.18 payoff ratio means mathematical expectancy is fragile; any increase in loss frequency or size immediately deepens the net loss, as already evidenced by the 0.94 PF.
- 【Execution Feasibility】Friction Sensitivity on XAUUSD H1: A thin average win of 4.76 (on 0.01 lots) cannot be verified against real‑world spread widening and slippage (test slippage set to 5); without live execution proof, the modest gross edge may not survive institutional frictions.
客观数学打分明细
- 【Baseline】Starts at 50 pts neutral baseline
- -20 pts / Adverse Grid Averaging: Adding positions at regular pip intervals against price movement
- -20 pts / Severely Unfavorable Payoff Ratio: Realized payoff ratio below 0.6, single large losses erode gains
- -20 pts / Net Loss: System incurred net loss of -0.83%
- -15 pts / Negative Expectancy Post-Outliers: PF drops strictly below 1.0 (0.83) after excluding top 5% trades, turning into net loss without windfall luck
- +15 pts / Minimal Drawdown (<10%)
- Final Strategy Safety Score: 0 pts (Grade E).
下一步行动与优化建议
- Export the full MT5 trade history with opening tickets, entry timestamps, and order comments to verify grid step spacing, basket composition, and the timing of recovery‑sizing events.
- Re‑run the backtest with a conservative cost model (active commission, swap, and realistic slippage) to test whether the 0.94 PF and 0.18 payoff remain materially unchanged or deteriorate further.
- Isolate the 'Smart Recovery' parameter effect: compare tested performance with recovery disabled versus the reported 1.6/3 configuration to quantify the incremental downside contribution.
- Conduct a forward/demo validation on a live feed of at least six months to observe slippage, spread widening, and actual fill behavior on XAUUSD H1.
- Define and test a hard per‑trade stop‑loss and a maximum grid‑level limit, then re‑evaluate the payoff ratio and post‑outlier expectancy under those constraints.
关于本报告的关键质询与答疑 (Q&A About This Report)
Why did this strategy receive Grade E despite an 84.16% win rate?
The grade is driven by internal downside metrics: a net loss of −83.04 (−0.83%), a realized payoff ratio of 0.18 (average win 4.76 vs average loss −27.02), and residual net profit of −217.24 with PF 0.83 after removing the top 5% of trades. The high win rate is overwhelmed by a small number of large losses (e.g. −41.40, −36.81), so expectancy is negative and outlier‑dependent.
The maximum drawdown is only 3.6%, which looks modest. Why is the strategy still rated poorly?
A 3.6% maximum drawdown is within the lowest risk band and is credited. However, the score is primarily penalized by negative net profit, a payoff ratio of 0.18, and post‑outlier net loss; capital preservation cannot offset a structurally loss‑making expectancy.
Several same‑second duplicate P/L rows appear (e.g. two rows at 2025.01.23 11:27:40 with identical 5.69). Does this confirm a basket or grid structure?
Equal P/L values at the same close timestamp are consistent with multiple positions or partial closes, but the supplied rows lack entry‑ticket pairing, so the exact structure (basket vs partial close) cannot be confirmed. The grid/recovery classification is based on the 'Smart Recovery 1.6 3' setting and the clustered loss pattern, not solely on duplicate rows.
What would change the Grade E assessment?
The verified record shows negative expectancy even before excluding outliers, so a change would require a materially different risk profile: a realized payoff ratio above 0.6, positive net profit after removing the top 5% of trades, and controlled loss sizing under recovery conditions. These are not present in the current data.
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.