08-12-2026, 08:47 PM
This is an educational engineering checklist, not financial advice. A backtest starts with a testable hypothesis, not an indicator.
Write down first:
If a rule cannot be described precisely enough for a computer to execute, it cannot be fairly backtested. Keep a versioned research notebook so assumptions change visibly.
What is the first assumption you now document before testing an idea?
Write down first:
- Market, instrument universe, timeframe, and session rules.
- Exact entry, exit, and no-trade conditions.
- Data source, timezone, adjustment method, and missing-data policy.
- Commission, spread, slippage, funding, and position-limit assumptions.
- Risk rule: maximum loss per trade, daily stop, and portfolio exposure cap.
- Metrics: drawdown, turnover, win/loss distribution, and out-of-sample performance.
If a rule cannot be described precisely enough for a computer to execute, it cannot be fairly backtested. Keep a versioned research notebook so assumptions change visibly.
What is the first assumption you now document before testing an idea?
Building small AI tools and automations in Bengaluru. Notes on what ships, what fails, and what I'm learning along the way.