Quantum Emperor V7.5 EA… AI Review | QuantTester
Quantum Emperor V7.5 EA · Me: return -0.8%, DD 3.43%, PF 0.94, 303 trades. Read the review.
Strategy Safety Grade: B- (55 / 100) · Caution Advised
Trading risk analysis result
Classified as Hybrid (Tiered Basket + Trend Following) per the precomputed ledger's lot-discipline context, the strategy is a high-win-rate (84.16%) mean-reversion scalper on XAUUSD H1 whose tight 3.6% max drawdown is offset by a negative expectancy (-0.83% over 632 days) and an average winner (4.76) far smaller than the average loser (-27.02), leaving Grade B- (55/100).
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。
Data transparency & Environment
- MT4/MT5: MetaTrader 5
- Symbol: XAUUSD
- Timeframe: H1 (2025.01.03 - 202
- Net profit: $-83.04
- Return: -0.8%
- Max drawdown: 3.43%
- Total Trades: 303
- Profit Factor (PF): 0.94
Critical Algorithmic Vulnerabilities
- Outlier Dependency (Return Consistency): The top 5% of trades (16 rows) contribute 11.05% of gross profit; removing them leaves residual net profit at -217.24 with PF 0.83, demonstrating that expectancy is driven by a thin set of statistical outliers rather than a durable systematic edge.
- Negative Expectancy & Payoff Asymmetry (Loss Aversion): With an 84.16% win rate and 0.94 PF, average winners are only 4.76 versus average losers of -27.02 (payoff ratio 0.18); modest choppy or trend-abort regimes rapidly flip arithmetic expectancy negative and the edge has not yet compensated accumulated losses over 632 days.
- Momentum Decay / Broken Quote Shape (D-Score): Peak balance near 10,169 in mid-March 2025 was followed by a protracted 3.6% slide to 9,916.96 by 2026-10-08, with no new equity high for the remaining 18 months, indicating alpha decay and a stagnant underwater trajectory rather than active recovery.
- Loss-Scale Escalation (Risk Stability): Mid-to-late sequence losses cluster between -26 and -41 (e.g. -36.81 on 2026-02-04, -41.40 on 2026-02-17, -33.48 on 2026-03-23), significantly wider than the early -25 to -26 baseline, indicating expanded stop distances or increased dollar exposure that undermines replication stability.
- Execution Friction Sensitivity (Capacity & Friction): Average holding of 125.4 minutes on H1 with no verified live broker execution; the modest realised profit factor of 0.94 and average win of 4.76 per 0.01 lot cannot be assumed to withstand live spread widening and slippage without forward evidence.
Score Breakdown Details
- [Base Score] All valid statements start at 50 points neutral benchmark (Grade B-)
- +15 pts [Lot Discipline]: Fixed/proportional lot execution, no adverse counter-trend doubling
- -20 pts [Negative Expectancy]: Net loss of -0.83% incurred, negative return
- -15 pts [Windfall Reliance]: Removing top 5% outlier deals drops system into loss (Remaining PF 0.83 < 1.0)
- +15 pts [Minimal Drawdown]: Max historical drawdown only 3.6% (< 10%), excellent downside preservation
- +10 pts [Sample Horizon]: Sample duration spans 632 days (>= 180 days), covering diverse regimes
- Final Strategy Safety Score: 55 / 100 (Grade B-).
Recommended Next Actions
- Rerun the backtest with the broker's actual spread and commission model on XAUUSD to test whether the 0.94 profit factor survives realistic cost increases.
- Extract and verify paired entry/exit tickets and direction for the 303 closed rows to confirm the tiered basket structure before allocating capital.
- Isolate and replay the top-5% outlier trades against real market data to determine whether their profit source is repeatable, and stress the strategy without them.
- Run a minimum 3-month demo forward test with live-equivalent execution to validate the 84% win rate and 0.18 payoff ratio outside the tester.
- Analyse the mid-to-late sequence loss escalation (-26 to -41) to confirm whether stop distances have drifted or absolute dollar exposure has grown, and whether SmartRecovery was triggered.
Q&A About This Report (Quant Due Diligence)
The strategy wins 84.16% of its 303 trades yet still loses 0.83%; how is that mathematically possible?
Payoff asymmetry. Average winners are 4.76 while average losers are -27.02, a 0.18 payoff ratio. Even with 255 wins and 48 losses, gross profit of 1,214.15 is outweighed by gross loss of 1,297.19, producing a 0.94 profit factor and an expectancy of -0.27 per trade. The high win rate does not compensate for the size of the 48 losses.
Why does the strategy still get Grade B- when the backtest shows a negative return?
The B- grade (55/100) reflects a balance of internal metrics: negative expectancy (-0.83%) and windfall reliance (residual PF 0.83 after excluding the top 5%) are the main deductions, offset by rigid 0.01-lot discipline (+15), a minimal 3.6% max drawdown (+15), and a 632-day sample (+10). The grade is not rescued by any single metric; it is a composite of real weaknesses and genuine capital-preservation strengths.
The 3.6% maximum drawdown looks low; does that mean the strategy is safe to allocate to?
No. A 3.6% max drawdown is objectively small and earns positive credit for downside preservation, but it coexists with negative net expectancy and an average loss 5.7 times the average win. The drawdown ceiling is a fact, not a guarantee of profitability. The absence of a live record, paired entry tickets, and cost-adjusted forward results means the low drawdown cannot yet be assumed to persist under real slippage.
Several losses cluster around -25 to -26 and later around -26 to -41; is that a fixed or widening stop?
The 48 loss rows are 85.4% clustered with a max/median ratio of 1.57, consistent with a consistent termination level, not an unbounded runaway. The later -36.81, -41.40, and -33.48 losses widen the absolute range but do not cross the unbounded-tail threshold; this is escalation within a controlled stop regime, not toxic holding. The exact stop mechanism (hard, trailing, or signal exit) cannot be confirmed from close-only rows.
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.