Sky City EA · MetaTrader 5 XAUUSD… AI Review | QuantTester
Sky City EA · MetaTrader 5 X: return 5,008.3%, DD 41.62%, PF 2.05, 70029 trades. Read the review.
Strategy Safety Grade: C (35 / 100) · High Risk Alert
Trading risk analysis result
Classified as Hybrid (Tiered Basket + Trend Following) with an overall PF of 2.05 but a 41.6% max drawdown and post-exclusion PF of 0.99; the strategy is a layered basket system whose edge is heavily concentrated in a small number of outlier trades, and it failed to demonstrate durable risk control.
Darwinex 集中度压力测试 (Top 5% Windfall Dependence Test)
剔除排名前 3502 笔（5%）极端盈利单后：毛利占比 51.68%，剩余净利润 $-4,478.70，剩余 PF 0.99。
Data transparency & Environment
- MT4/MT5: MetaTrader 5
- Symbol: XAUUSD
- Timeframe: H1 (2025.01.03 - 202
- Net profit: $500,829.60
- Return: 5,008.3%
- Max drawdown: 41.62%
- Total Trades: 70029
- Profit Factor (PF): 2.05
Critical Algorithmic Vulnerabilities
- Outlier Dependency: The top 5% of trades (3,502 rows) account for 51.68% of gross profit; removing them leaves residual net profit at -$4,478.70 with a PF of 0.99, demonstrating that expectancy is driven by statistical outliers rather than a durable systematic edge.
- Extreme Drawdown: Maximum equity drawdown reached 41.62% (from a peak balance of ~$510,829 to a trough), eroding nearly half of account equity at the worst point and indicating weak defensive execution.
- Loss Tail Risk: The largest single loss was -$1,412.40 and the stop_loss_evidence shows an unbounded_tail pattern (max_over_median 72.2), with 2.3% of losses clustered, indicating potential stop-loss widening or unprotected positions during adverse moves.
- Payoff Asymmetry Fragility: With a 63.8% win rate but average win of $21.91 and average loss of -$18.81 (payoff ratio 1.16), the strategy relies on a high win rate; any deterioration in win rate or larger outlier losses would quickly flip expectancy negative.
Score Breakdown Details
- [Base Score] All valid statements start at 50 points neutral benchmark (Grade B-)
- +5 pts [Order Basket]: Controlled multi-entry execution in same minute; layered basket architecture
- +5 pts [Expectancy]: Overall Profit Factor reached PF 2.05 with solid positive mathematical expectancy
- -15 pts [Windfall Reliance]: Removing top 5% outlier deals drops system into loss (Remaining PF 0.99 < 1.0)
- -20 pts [Extreme Drawdown]: Max historical drawdown reached 41.6% (> 35%), weak defensive execution
- +10 pts [Sample Horizon]: Sample duration spans 426 days (>= 180 days), covering diverse regimes
- Final Strategy Safety Score: 35 / 100 (Grade C).
Recommended Next Actions
- Conduct a sensitivity analysis on the top 5% of trades to quantify the impact of removing outliers on overall returns.
- Implement stricter stop-loss or position-sizing rules to reduce the maximum drawdown below 20%.
- Run a forward test on a demo account to validate performance under live spread and slippage conditions.
- Analyze the loss distribution to identify if stop-loss widening occurs during high-volatility events.
- Review the layered basket entry logic to ensure it does not inadvertently scale into losing positions without proper risk controls.
Q&A About This Report (Quant Due Diligence)
Why does the strategy have a Grade C despite a Profit Factor of 2.05?
The Grade C reflects the strategy's internal downside risk metrics. While the overall Profit Factor is 2.05, removing the top 5% of trades drops the residual Profit Factor to 0.99, showing that nearly all profit is dependent on a small set of outlier trades. Additionally, the maximum equity drawdown reached 41.62%, indicating significant capital impairment risk. These factors collectively justify the C grade.
How does the strategy handle stop-losses, and what does the evidence show?
The stop_loss_evidence shows a pattern of 'unbounded_tail' with a max_over_median of 72.23, meaning losses are not consistently terminated at a fixed level. The largest single loss was $1,412.40, and the clustered share of losses is only 2.3%. This suggests that while some stop-losses may be present, there is evidence of incomplete or widening stop protection during adverse moves.
What is the significance of the tiered basket entries observed in the trade rows?
Multiple orders opened in the same minute (e.g., 2025-01-09 10:26, 2025-02-03 11:16) indicate a layered basket execution style. This is common for strategies that seek to average into positions or capture short-term moves. However, without paired ticket evidence, the exact correlation and risk management of these entries cannot be fully confirmed.
Does the backtest provide any evidence of live viability?
This is a backtest report; no live or demo forward results are provided. The 426-day sample period and 70,029 trades offer substantial historical evidence, but real-world execution factors such as slippage and spread widening cannot be assessed from this data. Forward testing is necessary to confirm whether the backtested performance survives live market frictions.
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.