This is a fork of the original backtrader - slimmed down. Unnecessary features and updating the package to ensure compatibility with newer Python versions and dependencies.

Aims of this project:

• Slim down unnecessary features
• Improve performance
• Update aged implementations

For now the focus is:

• Code clean-up: remove unncessary imports
• Syntax update: make it more modern
• Remove deprecated integrations (i.e. pyfolio, IbPy, comtypes)
• Remove interactive plotting: the backend is heavy and slow
• Improved support for parallel processing

This is an ongoing process that has just started and will hopefully bring life to an excellent project.

Feel free to contribute!

A Python-based platform for live trading and backtesting, featuring:

• Live Data Feed and Trading:

  • Interactive Brokers (requires IbPy, significantly benefits from installed pytz)
  • Visual Chart (requires fork of comtypes until pull request integration, benefits from pytz)
  • Oanda (requires oandapy, REST API only – v20 streaming not supported)

• Data Sources:

  • CSV/files, online sources, or via pandas and blaze

• Data Management:

  • Filters (e.g., daily bars into intraday chunks, Renko bricks)
  • Multiple data feeds and strategies supported
  • Multiple simultaneous timeframes
  • Integrated resampling and replaying capabilities

• Backtesting Modes:

  • Step-by-step execution or all-at-once (strategy evaluation exception)

• Indicators:

  • Extensive built-in indicators (full list available here)
  • TA-Lib integration (requires Python ta-lib)
  • Easy creation of custom indicators

• Analyzers and Utilities:

  • Built-in analyzers (e.g., TimeReturn, Sharpe Ratio, SQN)
  • pyfolio integration (deprecated)

• Broker Simulation:

  • Supports multiple order types: Market, Close, Limit, Stop, StopLimit, StopTrail, StopTrailLimit, OCO, bracket orders, slippage, volume filling strategies, continuous cash adjustments for futures-like instruments

• Automated Staking:

  • Sizers for position sizing

• Cheating Modes:

  • Cheat-on-Close
  • Cheat-on-Open

• Schedulers and Calendars

• Plotting (requires matplotlib)

Backtrader is self-contained with minimal external dependencies (plotting requires matplotlib).

Currently, the installation takes place by navigating to the clone of this repository and running:

pip install -e

Works with:

• Python version >= 3.10

• Original backtrader repository: https://github.com/mementum/backtrader
• Blog: Backtrader Blog
• Docs: Full Documentation
• Indicators Reference: List of Built-in Indicators (122)

Follows format X.Y.Z.I where:

• X: Major version (stable, unless significant overhauls, e.g., numpy integration).
• Y: Minor version (new features or incompatible API changes).
• Z: Revision updates (documentation tweaks, minor changes, bug fixes).
• I: Number of built-in indicators.