Did your setup actually work? Backtesting
Test a strategy on history before you risk a rupee on it.
Every setup *sounds* good. The only way to know if one actually has an edge is to test it against history — that is backtesting. It is the difference between "I think buying RSI dips works" and "buying RSI(14) < 30 in stocks above their 200-day average returned X across Y trades."
What a backtest tells you
- •Win rate — how often the setup was profitable.
- •Average return per trade, and the total.
- •Number of trades — a 70% win rate over 6 trades is noise; over 300 it means something.
- •Drawdown — the worst losing stretch, so you know what you would have had to sit through.
How to think about the result
A backtest is a reality check, not a crystal ball. The past does not repeat exactly, and it is easy to fool yourself by tuning a strategy until it looks perfect on old data — 'curve-fitting' — which then falls apart live. Favour simple rules that work *okay* across many stocks and years over complex ones that look flawless on a handful.
Used honestly, backtesting kills your bad ideas cheaply — before the market does it expensively.
Tip · Backtest the setups from these lessons. If a rule can't show an edge over a few hundred historical trades, don't trade it with real money.
The full backtest engine + strategy library is part of Pro. See more →
AlphaGrid Learn is educational content, not investment advice.