MonkeyBeat

Can random stock picks beat professional fund managers?

MonkeyBeat generates equi-weighted portfolios of 10 randomly selected stocks from NSE indices and compares their performance against the underlying benchmark.

Test the "monkeys throwing darts" hypothesis yourself!

Quick Start

# Requires Python 3.12+ and uv
uv run app.py

Open http://localhost:7777

First run takes 5-10 minutes to fetch 5 years of stock data from Yahoo Finance.

Features

• Single file app - Everything in app.py (~850 lines)
• Zero setup - Embedded DuckDB, no external database
• 9 NSE indices - From NIFTY 50 to Smallcap 250
• Monte Carlo simulations - 100 random portfolios per request
• Fast - ~150ms response after data is loaded
• Shareable - UUID-based links with social sharing

Supported Indices

Index	Stocks	Category
NIFTY 50	50	Large Cap
NIFTY Next 50	50	Large Cap
NIFTY 100	100	Large Cap
NIFTY 200	200	Large Cap
NIFTY 500	500	Large Cap
NIFTY Midcap 100	100	Mid Cap
NIFTY Midcap 150	150	Mid Cap
NIFTY Smallcap 100	100	Small Cap
NIFTY Smallcap 250	250	Small Cap

How It Works

1. Select an index from the dropdown
2. App randomly picks 10 stocks from that index
3. Calculates returns over 1M, 6M, 1Y, 3Y, and 5Y periods
4. Compares against the benchmark index
5. Runs 100 Monte Carlo simulations to show win rate

Project Structure

monkeybeat/
├── app.py                    # FastAPI app (uv inline dependencies)
├── templates/                # Jinja2 templates
│   ├── base.html            # Layout with HTMX, Tailwind, Lightweight Charts
│   ├── portfolio.html       # Results page
│   └── error.html
├── scripts/
│   ├── download_indices.sh  # Fetch latest index constituents
│   └── ind_*.csv            # Index constituent lists
├── static/                   # Logo, favicon
└── data/                     # DuckDB database (auto-created)

Tech Stack

• FastAPI - Web framework
• DuckDB - Embedded analytics database
• yfinance - Yahoo Finance data
• HTMX - Dynamic HTML
• Tailwind CSS - Styling
• Lightweight Charts - TradingView charts
• uv - Python package manager

Updating Index Data

./scripts/download_indices.sh

Why Does the Monkey Usually Win?

1. Equal weighting - Your portfolio weights each stock equally, while indices are cap-weighted
2. Survivorship bias - Current index members are "survivors"; poor performers were removed
3. Concentration - 10 stocks can get lucky with 1-2 multibaggers

License

MIT

Credits

• Logo by @hashfyre

Disclaimer

This is for educational/entertainment purposes only. Not financial advice. Past performance does not guarantee future results.