📊 BackTesting Signals

A framework for extracting, backtesting, and optimizing cryptocurrency trading signals from Discord/Telegram channels.

---

🎯 What It Does

1. Extracts trading signals from Discord and Telegram channels
2. Backtests signals against real Binance historical price data
3. Optimizes performance by analyzing patterns (day/hour/coin combinations)
4. Generates actionable trading strategies with high win rates

---

⚡ Quick Start (5 minutes)

Prerequisites

• Python 3.10+
• Discord/Telegram API credentials
• Internet access (for Binance data)

Installation

# Clone repository
git clone https://github.com/yourusername/BackTestingSignals.git
cd BackTestingSignals

# Run setup script
python setup.py

The setup wizard will:

• Create virtual environment
• Install dependencies
• Guide you through configuration

Configuration

Edit config/config.json:

{
  "telegram": {
    "api_id": "YOUR_API_ID",
    "api_hash": "YOUR_API_HASH",
    "phone_number": "+1234567890",
    "channels": [{"name": "DaviddTech", "username": "DaviddTech"}]
  },
  "discord": {
    "token": "YOUR_DISCORD_TOKEN",
    "channel_id": "YOUR_CHANNEL_ID"
  }
}

Guides: docs/setup/telegram_setup.md | docs/discord-token-guide.md

---

🚀 Usage

1. Extract Signals

# Activate virtual environment first!
.\venv\Scripts\Activate.ps1  # Windows
source venv/bin/activate     # Linux/Mac

# Extract from Telegram (DaviddTech)
python extract_telegram.py

# Extract from Discord (Meta Signals)
python extract_discord.py

Signals saved to data/signals/

2. Run Backtest

python full_backtest.py

Select a signal file when prompted. Results saved to data/backtest_results/

3. Optimize

python long_short_optimization.py

Analyzes patterns to find high win-rate trading conditions.

---

📁 Project Structure

BackTestingSignals/
├── extract_telegram.py      # Telegram signal extraction
├── extract_discord.py       # Discord signal extraction
├── full_backtest.py         # Backtesting engine
├── long_short_optimization.py # Optimization analysis
├── analyze_davidtech.py     # DaviddTech analysis
├── setup.py                 # Setup wizard
│
├── src/                     # Core modules
│   ├── parsers/            # Signal format parsers
│   ├── backtesting/        # Backtesting logic
│   ├── data/               # Data management
│   └── analytics/          # Analysis tools
│
├── config/                  # Configuration
│   └── config.json         # API keys (DO NOT COMMIT)
│
├── data/                    # Data storage
│   ├── signals/            # Extracted signals
│   ├── backtest_results/   # Backtest outputs
│   └── cache/              # Binance data cache
│
├── docs/                    # Documentation
│   ├── setup/              # Setup guides
│   └── analysis/           # Analysis reports
│
└── archive/                 # Archived extraction methods

---

📊 Signal Sources Comparison

Source	Signals	Baseline WR	Profit Factor
Telegram (DaviddTech)	743	47.5%	1.33 ✅
Discord (Meta Signals)	1756	36.5%	0.93

Recommendation: DaviddTech provides better baseline performance.

Optimized Results (filtered strategies)

Source	LONG WR	SHORT WR
Telegram	75.0%	100.0%
Discord	80.8%	65.4%

---

🔬 How It Works

Backtesting Logic

For each signal:

1. Fetch 72 hours of 1-minute Binance OHLCV data
2. Track if price hits Target 1-3 or Stop Loss first
3. Record outcome, timing, and profit/loss

Optimization

1. Group signals by day, hour, coin, month
2. Calculate win rates for each group
3. Identify high-performance combinations (>60% WR)
4. Generate actionable trading rules

---

⚠️ Key Insights

The Thursday Curse 🚨

Both sources show poor performance on Thursday. Consider skipping Thursday signals.

Best Patterns (DaviddTech)

• LONG: Thursday, 11/15/19:00 UTC, LINKUSDT
• SHORT: Sunday, 03/08/21:00 UTC

Risk Management

• Maximum 2% risk per trade
• Always use stop losses
• Maximum 3 simultaneous positions

---

🛠️ Scripts Reference

Script	Purpose
extract_telegram.py	Extract signals from Telegram
extract_discord.py	Extract signals from Discord
full_backtest.py	Run backtests on signal files
long_short_optimization.py	Optimize LONG & SHORT strategies
analyze_davidtech.py	Comprehensive DaviddTech analysis
short_optimization.py	SHORT-only optimization
compare_long_short.py	Compare LONG vs SHORT
check_telegram_channel.py	Verify Telegram access

---

🐛 Troubleshooting

Token/API Issues

• Expired token: Re-extract from browser
• Rate limit: Built-in retry handles this
• No data: Check symbol format (BTCUSDT not BTC-USDT)

Installation Issues

• Requires Python 3.10+
• Delete venv/ and re-run setup.py if venv issues

---

📄 License

MIT License

---

Version: 2.2
Last Updated: December 2025
Status: Production Ready ✅