- Was the rule fixed before you saw results?
Rules tuned after the fact fit the past, not the future.
- Could the rule have known this at the time?
Using information that wasn't available yet is look-ahead bias.
- Are trading costs included?
Commission, taxes, spread and slippage can erase a small edge.
- Are delisted companies included?
Testing only today's survivors flatters results.
- Are prices adjusted for bonus shares, splits and dividends?
Unadjusted prices create false drops, common on the PSX.
- Did you keep some data unseen?
Test once on a held-back period. A collapse there signals overfitting.
- How many versions did you try?
Try enough and one will look good by chance.
- What is the benchmark?
Compare with simply holding the index.
- How bad was the worst fall?
Maximum drawdown shows what you'd have had to sit through.
- Does it survive small changes?
If a tiny tweak breaks it, be cautious.
Educational content only. Not investment advice.