How Many Years of Data Should a Backtest Cover?
More is not simply better. What a given span of history can and cannot tell you, and why what the period contains matters more than how long it is.
Read articleMore is not simply better. What a given span of history can and cannot tell you, and why what the period contains matters more than how long it is.
Read articleEach of these produces a backtest that looks better than reality, and each is easy to make without noticing. How to spot them in your own results.
Read articleEach stage answers a question the previous one could not. Skipping a stage does not speed things up — it just moves the discovery to where it costs money.
Read articleWhere free backtesting tools are genuinely useful, the five limits that usually appear, and how to tell whether a free result is good enough to act on.
Read articleCurve fitting is tuning a strategy until it describes one past rather than any real market behaviour. How it happens, the tests that detect it, and the process changes that prevent it.
Read articleSlippage, assumed fills, costs, survivorship in your own testing and regime change — the structural reasons a backtest overstates live performance, and how to narrow the gap.
Read articleMax drawdown, expectancy, reward-to-risk, loss streaks and return-over-max-drawdown explained — and the order to read them in so the headline profit number does not mislead you.
Read articleA step-by-step guide to backtesting NIFTY and SENSEX option strategies — what to test, how long a sample you need, which metrics matter, and the mistakes that make a backtest lie to you.
Read article