BacktestBase Documentation
Complete guide to using BacktestBase for systematic trading strategy analysis and portfolio optimization
Table of Contents
What is BacktestBase?
BacktestBase is the comprehensive platform for systematic traders who want to transform their individual TradingView strategies into a professional trading foundation. Unlike simple analysis tools, BacktestBase serves as your central hub for strategy organization, portfolio creation, and risk management.
Key Benefits:
- Strategy Organization: Automatically organize all your TradingView backtests in one searchable database
- Portfolio Creation: Combine multiple strategies into diversified portfolios with scientific weighting
- Risk Analysis: Run Monte Carlo simulations using your actual strategy performance data
- Performance Tracking: Compare strategies side-by-side with comprehensive metrics
- Professional Foundation: Build a systematic approach to trading with data-driven insights
Account Setup
Getting Started
- Sign Up: Create your account using Google OAuth - secure and instant
- Dashboard Access: Once signed in, you'll be redirected to your personal dashboard
- First Upload: Start by uploading your first TradingView strategy export
Dashboard Overview
Your dashboard is the command center for all your trading strategies and analysis.
Dashboard Sections:
Quick Stats
- Total strategies uploaded
- Combined net profit
- Average win rate
- Best profit factor
Strategy Management
- Sort by any performance metric
- Edit strategy names and descriptions
- Delete outdated strategies
- View detailed analysis
Uploading TradingView Strategies
Step-by-Step Upload Process
1. Open Strategy Report
Click the "Strategy Report" tab at the bottom of your TradingView chart.
2. Open Strategy Dropdown
Click your strategy name dropdown to open the menu.
3. Download Data as XLSX
Select "Download data as XLSX" from the dropdown menu.
Do Not Open or Edit Before Upload
Upload the file DIRECTLY to BacktestBase without:
- Opening it in Excel, Numbers, or Google Sheets
- Renaming or moving the file
- Making any edits
Opening the file may corrupt data formatting and cause upload errors.
Do Not Use the CSV Download Button
When viewing the "List of trades" tab, do not use the download button on the right side of the panel. This only exports a CSV file which is incompatible with BacktestBase. Always use the strategy dropdown menu to download the XLSX file.
4. Upload to BacktestBase
- Go to the Upload page in BacktestBase
- Drag and drop your .xlsx file or click to browse
- Wait for automatic parsing
- Review the extracted data preview
- Add a descriptive name and optional description
- Click "Save Strategy" to add it to your database
Before Upload Checklist
- File extension is .xlsx (not .csv)
- Downloaded from strategy dropdown menu
- File NOT opened or edited after download
What Gets Extracted
- Net Profit ($ and %)
- Gross Profit/Loss
- Profit Factor
- Sharpe & Sortino Ratios
- Maximum Drawdown
- Complete trade history
- Win/Loss statistics
- Average winning/losing trades
- Largest winning/losing trades
- Long/Short performance breakdown
Important Notes
- Only TradingView .xlsx exports are currently supported
- Files must be under 10MB in size
- Free accounts limited to 3 strategies
- All data is encrypted and secure
Currency Handling
BacktestBase automatically detects and processes strategy currencies from your TradingView exports.
Supported Currencies (24)
USD, EUR, GBP, JPY, CHF, AUD, CAD, NZD, SEK, NOK, HKD, SGD, TRY, ZAR, BTC, USDT, ETH, RUB, MYR, KRW, INR, PKR, PLN, EGP
Display Currency vs Instrument Currency
TradingView may display values in your preferred display currency rather than the instrument's quote currency. BacktestBase uses the instrument's quote currency for accurate calculations.
If a mismatch is detected, an info banner will explain the difference between display and calculation currencies.
Mixed Currency Portfolios
When combining strategies with different quote currencies in a portfolio, total values may show asterisks (*) to indicate that values cannot be accurately summed without currency conversion. Individual strategy metrics remain accurate in their respective currencies.
RTL Language Limitation
Right-to-left languages (Hebrew, Arabic) are not supported due to a known issue with TradingView's XLSX export formatting. Please use TradingView with English or another left-to-right language for exports.
Strategy Analysis
BacktestBase provides comprehensive analysis of your trading strategies with professional-grade metrics and insights that go beyond basic TradingView statistics.
Available Analysis:
Performance Metrics
- Profit Factor & Risk-Adjusted Returns
- Sharpe, Sortino, and Calmar Ratios
- Maximum & Average Drawdown Analysis
- Win Rate & Risk/Reward Ratios
- Long vs Short Performance Breakdown
Trade Analysis
- Complete trade history with P&L
- Average P&L per trade
- Average winning/losing trade sizes
- Largest winning/losing trades
- Long vs Short directional performance
Strategy Comparison
Use the dashboard sorting and filtering features to compare strategies across different:
- Market symbols (EURUSD vs GBPUSD vs crypto)
- Timeframes (15m vs 1h vs 4h strategies)
- Strategy types (trend-following vs mean-reversion)
- Risk profiles (high Sharpe vs low drawdown)
Portfolio Creation
Transform individual strategies into diversified portfolios using professional weighting algorithms and concrete benefit analysis.
Portfolio Weighting Methods
Account Size Weighting
Weight strategies based on their original backtest capital allocation. Best for realistic portfolio representation when strategies were tested with different account sizes.
Equal Weight Distribution
Simple 1/N allocation across all selected strategies. Provides maximum diversification and is ideal when you want equal exposure to each trading approach.
Risk-Adjusted Weighting
Allocate more capital to strategies with lower maximum drawdown. Optimizes for risk-adjusted returns and smoother portfolio performance.
Portfolio Benefits Analysis
Capital Efficiency
Compare total portfolio capital requirements vs individual strategy needs.
Risk Reduction
Measure portfolio drawdown improvement vs worst individual strategy.
Trading Consistency
Analyze increased trading opportunities through combined signals.
Market Coverage
Evaluate diversification across symbols and timeframes.
Monte Carlo Stress Testing
Research-Backed Methodology: BacktestBase uses institutional-grade Monte Carlo stress testing for individual strategy analysis, providing superior accuracy through pure mathematical trade randomization and skipping methodologies based on established statistical principles.
Our Monte Carlo system evaluates individual strategy robustness using actual historical trades - no forward projections or statistical assumptions, just rigorous testing of real performance data under various execution scenarios.
Scientific Foundation
Historical Bootstrap Method
Based on Efron & Tibshirani (1993) bootstrap theory
- Uses actual trade sequences, not synthetic distributions
- Eliminates parametric assumptions
- Superior to normal distribution models
Block Bootstrap Technique
Advanced implementation of Politis & Romano (1994)
- Preserves temporal correlation structures
- Captures regime-dependent relationships
- More realistic than IID sampling
Technical Implementation
Enhanced Strategy Identity Preservation:
- Trade Extraction: Parse all historical trades from TradingView exports maintaining chronological order
- Trade Skipping: Each simulation skips exactly the user-defined percentage of trades (randomly selected) to simulate missed opportunities
- Sequence Randomization: Remaining trades are randomly shuffled to test robustness across different timing scenarios
- Peak-to-Trough Analysis: Calculate drawdowns using industry-standard peak capital methodology on final sequences
- Robustness Scoring: 30-point system based on drawdown control and profit consistency across all simulations
Advantages Over Traditional Methods
Superior Accuracy
- Uses actual trade data, not synthetic distributions
- Captures real strategy behavior patterns
- No parametric assumptions about returns
- Works with any instrument or market
Technical Superiority
- Bootstrap methodology for robust statistics
- Eliminates IID assumptions
- Preserves temporal structures
- Pure mathematical approach
Key Parameters & Interpretation
Simulation Count
500-10,000 independent runs
Higher counts increase statistical significance
Trade Skip Percentage
Exact user-defined percentage per simulation (1-50%)
Simulates realistic execution gaps
Sequence Randomization
Remaining trades randomly shuffled
Tests robustness across different timing scenarios
30-Point Robustness Scoring System
Uses smooth linear interpolation between anchor points for nuanced scoring.
Applied to 5th, 50th, 95th percentiles (5 pts each)
- 0% drawdown → 5.0 points
- 25% drawdown → 3.0 points
- 50% drawdown → 1.0 point
- 75%+ drawdown → 0.0 points
- Linear interpolation between anchors
Ratio = Net Profit % ÷ Max Drawdown %
- Ratio 4.0+ → 5.0 points
- Ratio 3.0 → 4.0 points
- Ratio 2.0 → 2.5 points
- Ratio 1.0 → 0.5 points
- Ratio <1.0 → 0.0-0.5 points
Account Ruin Override: If median (50th percentile) drawdown exceeds 100%, the strategy automatically receives an F grade with all scores zeroed, regardless of other metrics.
Grades: A+ (28-30), A (25-27), B+ (22-24), B (19-21), C+ (16-18), C (13-15), D (10-12), F (0-9)
Practical Applications
Strategy Validation
- Test robustness under execution stress
- Identify strategies sensitive to timing
- Plan for realistic trading conditions
Portfolio Construction
- Compare weighting methods
- Evaluate diversification benefits
- Optimize position sizing
Risk Management
- Understand worst-case scenarios
- Plan capital requirements
- Set realistic expectations
Institutional Analysis
- Professional-grade stress testing
- Research-backed methodology
- Academic foundation compliance
Risk of Ruin Simulation
Beyond percentile analysis, BacktestBase calculates the probability of hitting catastrophic drawdown levels through simulation-based Risk of Ruin analysis.
Configurable Thresholds
- 25% account loss threshold
- 50% account loss threshold
- 75% account loss threshold
- 100% account loss (total ruin)
- Custom percentage threshold
Results Include
- Probability of hitting ruin threshold
- Median trades before ruin occurs
- Average trades before ruin occurs
- Lowest equity point reached
- Risk per trade used in simulation
How It Works
Risk of Ruin is calculated by tracking which Monte Carlo simulations hit the specified drawdown threshold. The system uses your configured risk per trade (or derives it from strategy metrics) to simulate thousands of possible equity curves and determine the probability of catastrophic loss under various market conditions.
Saving Stress Test Results
After running a Monte Carlo stress test, you can save the results directly to your strategy for future reference and comparison across your portfolio.
What Gets Saved
- Robustness grade (A+ through F)
- Robustness score (0-30 points)
- Percentile results (5th, 50th, 95th)
- P&L and drawdown distributions
- Risk of Ruin probability and metrics
Benefits
- View robustness grades on dashboard
- Compare robustness across strategies
- Track strategy robustness over time
- Quick filtering by grade
- Visible on strategy cards
Statistical Validity Panel
Beyond robustness grades, the Monte Carlo results include a Statistical Validity Panel with three complementary metrics that validate whether your backtest results can be trusted.
Returns Confidence
Uses a t-statistic to test whether positive returns are statistically significant or just random noise.
- What it tests: Are the returns genuinely positive, or could they occur by chance?
- Thresholds: 90%+ Likely accurate, 95%+ Verified, 99%+ Highly verified
- When it matters: Low confidence suggests more trades are needed to validate strategy edge
Win Rate Confidence
Uses a z-test to verify whether the observed win rate is statistically accurate.
- What it tests: Is a 60% win rate truly 60%, or could it be 50% with lucky variance?
- Works for: Both trend-following (40% win rate) and mean-reversion (60%+) strategies
- Interpretation: Tests deviation from 50% - a 40% win rate can be just as verified as 60%
Sharpe Validation — Min Track Record Length (MinTRL)
Based on Bailey & López de Prado's formula, calculates the minimum trades needed to validate your Sharpe ratio.
- What it shows: Minimum sample size to trust the strategy's risk-adjusted returns
- Formula basis: Accounts for return skewness and kurtosis (fat tails)
- Status: Verified (✓) when actual trades exceed minimum required
Validity vs Quality
Statistical validity answers "Can we trust these results?" - not "Is this a good strategy?" A strategy can have verified statistics but still be unprofitable. Conversely, a profitable strategy with uncertain validity needs more data before you can trust the edge is real.
Equity Probability Cone
Click the "Equity Curves" button in the Monte Carlo results to view a Kevin Davey-style probability cone visualization showing the range of possible equity paths from your simulations.
What the Cone Shows
- Extreme range (light gray): 5th to 95th percentile bounds
- Likely range (darker gray): 25th to 75th percentile bounds
- Stressed median (teal line): 50th percentile - expected outcome
- Original backtest (orange dashed): Where your actual result ended
How to Interpret
- Cone widens over time as uncertainty grows
- Original above median = backtest may have been lucky with trade timing
- Original below median = backtest had unlucky trade sequence
- Original near median = result is typical for this strategy
Methodology
The probability cone runs 1,000 dedicated simulations using the same skip + shuffle methodology as the main Monte Carlo stress test. Each simulation skips a percentage of trades (matching your selected skip rate) and randomizes the order of remaining trades. The percentile bands show the distribution of possible outcomes at each trade point.
Academic References
- Efron, B., & Tibshirani, R. (1993). An Introduction to the Bootstrap. Chapman & Hall/CRC
- Politis, D. N., & Romano, J. P. (1994). The Stationary Bootstrap. Journal of the American Statistical Association
- Bailey, D. H., & López de Prado, M. (2014). The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting and Non-Normality
- Getmansky, M., Lo, A. W., & Makarov, I. (2004). An Econometric Model of Serial Correlation and Illiquidity in Hedge Fund Returns
- Campbell, J. Y., Lo, A. W., & MacKinlay, A. C. (1997). The Econometrics of Financial Markets. Princeton University Press
Frequently Asked Questions
What file formats do you support?
Currently, BacktestBase supports TradingView Strategy Tester XLSX exports with full 5-sheet parsing. Support for additional platforms and formats is planned for future releases.
How secure is my trading data?
We use enterprise-grade security with Google OAuth authentication, database encryption, and Row Level Security. Your data is isolated to your account and never shared. We are not a broker - just a secure analysis platform.
Can I analyze strategies from different markets?
Absolutely. BacktestBase works with any TradingView strategy regardless of market (Forex, crypto, stocks, futures) or timeframe. Build comprehensive portfolios across multiple markets and trading approaches.
What's the difference between BacktestBase and TradingView?
TradingView is excellent for backtesting individual strategies. BacktestBase takes those results and transforms them into a comprehensive trading foundation - organizing strategies, creating portfolios, running risk simulations, and building systematic approaches to trading.
Can I delete my account and data?
Yes, you have full control over your data. You can delete individual strategies or your entire account at any time. Data deletion is permanent and immediate.
Need More Help?
Can't find what you're looking for? Our support team is here to help with any questions about BacktestBase features, strategy analysis, or portfolio optimization.
Response Time: We typically respond within 24 hours. Include your account email and describe your question in detail for the fastest response.