1. TL;DR: What can QuantTester tell you?
Trading report analysis converts historical order records into readable metrics and investigative forensic questions. QuantTester strictly distinguishes mathematically confirmed facts from unknown market variables; it does not authenticate legal account ownership, rerun codes on independent market feeds, or predict future profitability.
| Core Inquiries | What QuantTester Provides | Critical Boundaries |
|---|---|---|
| How did the report perform? | Net Profit, true Profit Factor (PF), Peak-to-Trough Drawdown, and full native MT5 metrics computed from 100% of closed trades. | Scope depends on raw file integrity and exported field coverage. |
| What hidden risks exist? | Deterministic deductions, Martingale doubling detection, disaster stops, Darwinex outlier stress tests, and AI forensic reconstruction. | 0–100 score reflects structural resilience and downside defense, not future profit odds. |
| Can I share and publish? | Generate encrypted private share links or publish qualifying reports to the public catalog. | Accessible to anyone with the secret token; sharing does not imply official endorsement. |
2. How are return, profit factor and drawdown calculated?
Return and Profit Factor compare profits with capital and losses, while drawdown exposes the deepest peak-to-trough decline. Balance and Equity series must be strictly distinguished.
Official Metric Definitions:
1. Profit Factor (PF): Gross profit divided by the absolute value of gross loss over the same period.
2. Maximum Relative Drawdown: The largest percentage decline from a preceding peak to a subsequent trough in the balance or equity curve.
Formulas adhere strictly to MetaTrader 5 Official Documentation. QuantTester reconstructs the balance curve directly from 100% of closed trade executions, preventing source reports from concealing true peak-to-trough drawdowns.
| Metric Name | Formula & Derivation Rule | Boundary & Scope |
|---|---|---|
| Total Return (%) | Net Profit ÷ Capital Base × 100% |
Uses initial deposit; cash flows must be interpreted separately. |
| Profit Factor (PF) | Gross Profit ÷ |Gross Loss| |
Undefined when gross loss is 0; computed across full closed trades. |
| Win Rate (%) | Profitable Closed Trades ÷ Total Closed Trades × 100% |
Break-even exits count in denominator; high win rate ≠ positive expectancy. |
| Payoff Ratio | Average Win ÷ |Average Loss| |
Measures per-trade win/loss symmetry; distinct from Profit Factor. |
| Max Relative DD (%) | (Peak Equity - Subsequent Trough) ÷ Peak Equity × 100% |
Reconstructed from execution stream, avoiding 0% report anomalies. |
3. How do backtests, demo accounts and live histories differ?
Backtests simulate historical ticks, demo accounts operate in forward sandboxes, and live accounts reflect real-money executions. No single static file can prove account ownership or data authenticity.
| Dimension | Strategy Tester Backtest | Demo-Account History | Live-Account History |
|---|---|---|---|
| Data Origin | Historical tick simulation | Broker simulated execution | Real-money broker execution |
| Execution & Costs | Tester spread/commission assumptions | No real-world liquidity slippage | Reflects live spread, swap & execution friction |
| Supported Insight | Logic robustness under assumptions | Forward stability in testbed | Live execution discipline & operational friction |
| Unknown Gaps | Curve-fitting & unknown regimes | Live liquidity shock & emotional stress | Statement tampering & cherry-picked intervals |
4. What does the 0–100 AI Strategy Safety Score mean?
The 0–100 Strategy Safety Score quantifies a trading system's underlying robustness, downside preservation mechanisms, and tail-risk defense; it is never a forecast of future gains.
Traditional rating portals use opaque continuous regression models. QuantTester breaks the black box with an open Deterministic Quantitative Ledger (50-Point Neutral Baseline + Discrete Evidence-Backed Debits & Credits).
All reports begin at 50 points (Grade B- Neutral Baseline). Every point added or deducted is mathematically verifiable in the raw closed-order records.
Benchmarked Against Darwinex D-Score Methodology
QuantTester's risk methodology draws heavily from the institutional standards of Darwinex's D-Score 12 Investable Attributes (IAs):
- Loss Aversion (LA) Audit: Measures holding duration asymmetry between winning and losing trades, detecting tendencies to cut profits prematurely while holding losses for days.
- Outlier Robustness & Return Quality: Eliminates the top 5% most profitable windfall trades to test residual expectancy. Profit must stem from repeatable edge, not lucky tail events.
- Open/Closing Risk (OS/RC): Verifies disciplined hard stop loss execution across trades and exposes "disaster stops" exceeding 20% of capital.
- Experience & Sample Size (EX): Audits with under 50 closed trades cannot mathematically exclude random noise; score is capped at 74 pts (Grade B+) to prevent curve-fitted high scores.
5. Hard Circuit Breaker: Forced Liquidation Floor (Stop Out)
[MANDATORY HARD CIRCUIT BREAKER]
If any trade record contains broker
close at stopforced liquidations, balance wipeouts, or negative balance adjustments, the system triggers an immediate circuit breaker, capping the score at 5 pts (Grade E Catastrophic) as a one-vote veto.Regardless of historical return multiples, margin stop-outs indicate fatal money management flaws, disqualifying the strategy as an investable asset.
6. 8 Quantitative Risk Dimensions & Ledger Rules
The system applies deterministic rules across 8 objective dimensions based on execution timestamps, lot sizes, stop losses, outliers, and duration ratios:
| Dimension | Objective Evidence & Rule Criteria | Points | Risk Rationale |
|---|---|---|---|
| 1. Sizing & Discipline |
• Fixed / proportional lot sizing without counter-trend averaging • Synchronized multi-order basket entry within same minute • Counter-trend grid adding at intervals (Adverse Grid) • Geometric lot multiplier scaling during floating loss (Martingale) |
+15 pts +5 pts -20 pts -35 pts |
Eliminates one-way breakout wipeouts; severely penalizes exponential Martingale lot explosion. |
| 2. Stop-Loss Defense |
• Disciplined hard stop loss executed across trades, bounding single-exit risk • Dynamic trailing stop or breakeven defense locking in positive floating profits • Stop loss present but exit > 20% or concurrent exit > 35% of capital (Disaster Stop) • No hard stop loss; toxic unhedged holding until deep drawdown (≥50%) |
+20 pts +10 pts +0 pts -25 pts |
Hard stops and trailing stops are verified independently; oversized stops receive no credit; unhedged holding is penalized. |
| 3. Expectancy & Payoff |
• Realized payoff ratio > 1.8 (asymmetric profit-to-loss edge) • Realized payoff ratio < 0.6 (excluding scalping / tiered baskets) • Full-sample profit factor PF ≥ 1.15 (fundamental positive mathematical expectancy) |
+15 pts -20 pts +5 pts |
Ensures robust statistical edge where average gains comfortably cover operational friction. |
| 4. Outlier Robustness (Darwinex D-Score) |
• Excluding top 5% windfall trades leaves residual PF ≥ 1.2 • Excluding top 5% windfall trades leaves residual PF ≥ 1.0 (breakeven preservation) • Excluding top 5% windfall trades turns performance negative (residual PF < 1.0) |
+10 pts +5 pts -15 pts |
Tests whether profitability is repeatable or dependent on lucky tail events; surviving outlier removal is an institutional gate. |
| 5. Drawdown Depth |
• Peak historical drawdown < 10% (requires ≥ 50 closed trades) • Peak historical drawdown 10%–20% (controlled institutional band) • Peak historical drawdown 20%–35% (elevated equity volatility) • Peak historical drawdown > 35% (fragile downside preservation) |
+15 pts +5 pts -10 pts -20 pts |
Measures observed equity stress and capital preservation tolerance. |
| 6. Disposition Effect (Loss Aversion) |
• Loss-to-win average holding duration ratio ≥ 4.0:1 with losses ≥ 2 hours • Loss-to-win holding duration ratio ≤ 1.2:1 with winning trades ≥ 10 minutes |
-15 pts +5 pts |
Identifies the behavioral bias of cutting profits prematurely while passively holding toxic floating losses. |
| 7. Calmar Ratio |
• Return-to-drawdown ratio ≥ 2.5 with maximum drawdown ≤ 25% • Maximum drawdown ≥ 30% with return-to-drawdown ratio < 0.6 |
+10 pts -10 pts |
Quantifies excess return generated per unit of downside equity risk endured. |
| 8. Sample Size Cap | • Fewer than 50 closed trades and raw score ≥ 75 pts: score automatically capped at 74 pts (Grade B+) | Cap (74 pts) | Statistical significance guard: small samples cannot rule out random luck or curve-fitting. |
7. 13 Discrete Safety Grade Tiers
The final score is strictly clamped to the [0, 100] interval and mapped to 13 discrete grades:
| Safety Grade | Score Band | Qualitative Risk Profile |
|---|---|---|
| Grade A+ | 90–100 pts | Institutional excellence: strict hard stops, minimal drawdown, superior risk-adjusted return. |
| Grade A / A- | 75–89 pts | Robust and resilient: mature risk controls, positive mathematical expectancy, controlled drawdown. |
| Grade B+ / B / B- | 50–74 pts | Viable standard execution: baseline defense (50 pts baseline = B-; sample cap = B+ 74 pts). |
| Grade C+ / C / C- | 20–49 pts | Structural vulnerabilities: elevated drawdown, outlier dependency, or unfavorable payoff ratio. |
| Grade D+ / D / D- | 6–19 pts | Severe downside fragility: lack of stop loss, negative expectancy, or acute holding bias. |
| Grade E | 0–5 pts | Forced liquidation (Stop Out), capital wipeout, or catastrophic unhedged failure (hard floor). |
8. Upload, Parsing & AI Review Workflow
Export your HTML/HTM statement from MetaTrader, drag it into QuantTester for instant non-blocking parsing, and optionally request a deep forensic review from our cloud quantitative LLM.
Step 1: Local In-Browser Worker Parsing
Import your report into QuantTester Auditor (UTF-8, UTF-16LE, UTF-16BE auto-detected). Execution parsing runs 100% locally inside your browser's Web Worker sandbox; raw files are never stored permanently on our servers.
Step 2: AI Quantitative Forensics
Click "Start AI Review" to transmit anonymized metrics and sampled key trades via TLS 1.3 encryption. The model reconstructs maximum drawdown episodes, cross-examines backtest friction, and generates a comprehensive diligence report.
Step 3: Private Sharing & Public Cataloging
Reviews generate private secret tokens by default (unindexed by search engines). Qualifying reports (single record, ≥30 trades, conclusions ≥40 chars) can optionally be published to the Public Reports Directory.