Backtest momentum strategies without hindsight
A backtest is only useful if it uses what you could have known at the time. Market Strength Scanner’s backtests are built to avoid the most common traps.
Open the free scannerWhat “point-in-time” means
Each month-end the test considers only stocks that were large and liquid then, using that month’s price and traded value, and ranks them using data up to that day. Trades happen at the next trading day’s close, not at the price that triggered the signal.
Three kinds of backtest
- Stock momentum: rank by a weighted mix of 1-year, 6-month, 3-month and 1-month returns, hold the top N and rebalance on a schedule, with optional buffers and market filters.
- Sector rotation: rank themes, buy the strongest themes’ strongest stocks, and compare with official NSE thematic indices.
- Rules-based: set entry, stop, target and trailing-exit rules and test them on daily data.
Costs, tax and what is still missing
Results include trading costs and can include Indian capital-gains tax (short- and long-term). Known limits are shown next to the numbers: delisted stocks are missing, share counts are today’s, and fills are idealised. Compare rule sets with each other rather than treating one number as a forecast.
Avoiding overfitting
Choose settings on one period and check them on another. A rule that only works for one narrow parameter value is more likely to be luck than edge.
Frequently asked questions
What is a point-in-time backtest?
A backtest that uses only information available on each historical date, including which stocks existed and their size then, so it does not benefit from knowing the future.
Are the results a forecast?
No. They show how a rule set would have behaved in the past, with known limitations. Future results can differ materially.
Do backtests include costs and tax?
Yes. You can set the cost per trade and turn tax on or off; the settings are shown beside the results.
Which markets can I test?
Indian NSE stocks.