Skip to content

Curated collection of top-performing trading strategies discovered through BART evolution runs

Notifications You must be signed in to change notification settings

gopher-lab/strategies

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

BART Top Trading Strategies

Curated collection of top-performing trading strategies discovered through BART (Backtest Analysis & Research Tool) evolution runs.

Performance Summary

Rank Strategy Symbol PnL Sharpe Profit Factor Max DD Win Rate Trades
1 Simplified Momentum Conservative TAO +104.19% 1.67 1.74 11.76% 41.03% 117
2 Enhanced Long-Only V4.1 TAO +93.77% 1.91 2.00 7.31% 41.56% 77
3 Ultra-Selective Range TAO +57.70% 1.70 2.05 12.68% 47.27% 55
4 Short-Focused Momentum TAO +45.17% 0.94 1.39 18.64% 36.19% 105
5 Moderate Range Trading BTC +12.79% 17.22 3.56 3.42% 66.67% 6

Backtest Parameters

  • Exchange: Hyperliquid
  • Leverage: 10x
  • Initial Balance: $10,000
  • TAO Test Period: Dec 22, 2025 - Jan 16, 2026 (25 days)
  • BTC Test Period: Jan 10-16, 2026 (6 days)
  • Model: qwen/qwen3-vl-8b-instruct

Usage

Each strategy folder contains:

File Description
strategy.md Human-readable strategy breakdown
prompt.txt Raw prompt for copy/paste into trading agents
backtest-results.json Full backtest data with all trades
metrics.json Quick performance metrics summary

Quick Start

  1. Navigate to a strategy folder
  2. Copy contents of prompt.txt
  3. Paste into your trading agent's system prompt
  4. Configure for your target symbol and timeframe

Strategy Selection Guide

Use Case Recommended Strategy
Maximum returns, moderate risk Simplified Momentum Conservative
Best risk-adjusted returns Enhanced Long-Only V4.1
Lowest drawdown BTC Moderate Range Trading
Highest win rate BTC Moderate Range Trading
Range-bound markets Ultra-Selective Range
Bearish bias Short-Focused Momentum

Notes

  • All strategies use 15m primary timeframe with multi-timeframe confirmation
  • Performance may vary based on market conditions
  • Always paper trade before live deployment
  • Consider reducing leverage for live trading

Contributing

When adding new strategies:

  1. Create folder under appropriate symbol directory
  2. Include all 4 required files
  3. Update this README with performance metrics
  4. Include original backtest session ID for traceability

About

Curated collection of top-performing trading strategies discovered through BART evolution runs

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published