.68% vs 47%: I Asked AI to Backtest a Simple Momentum Strategy
Summary
Highlights
Strategy Overview and Stock Selection00:00:02
The video introduces a system that beat the flat Nifty market by selecting 10 stocks from the Nifty 200 universe with the highest sharp ratio to ensure smooth returns.
The Trading Rules and Setup00:01:07
The strategy involves entering stocks with a 10% trailing stop loss and trailing it higher daily after market close, with stock selection done via Quantinus and backtested using Claude AI.
Version One Results00:02:13
Version one (buy and trail with cash sitting idle after stop loss) yielded a 20% return, turning 5 lakh rupees into 6 lakh rupees with only a 2.6% maximum drawdown.
Version Two Results00:02:51
Version two redeployed capital from stopped-out stocks into surviving winners, resulting in a 47% return, turning 5 lakh rupees into 7.37 lakh rupees with a 7.7% drawdown.
Critical Considerations and Conclusion00:04:04
The narrator warns that results vary by market regime, outperformance relied on a single outlier winner (Laurus Labs), and the backtest excludes taxes, brokerage, and slippage.
Recently Summarized Articles
Related summaries
- This ETF Momentum Strategy Feels Like Cheating (+39% CAGR Backtest)
- ETF Momentum Strategy: 23.9% CAGR Backtest (Simple Monthly Rules)
- I Backtested 1,000 Trading Strategies—Here's What Actually Matters
- We Backtested a SPY Strategy From 1998 to 2026
- Turn $10K into $430K with This Short Selling Strategy (Realtest Backtest)
- 7 Systematic Algo Trading Strategies (Backtests And Rules)