Treat the backtest as an experiment
Design a backtest that isolates factors, responses, and uncontrolled conditions before interpreting performance.
The move: design the test before you admire the result. Factors Factors are the inputs you deliberately vary: lookback, rebalance cadence, position cap, filter threshold, execution window, data vendor, and cost assumption. Keep the list short enough that each factor can be interpreted. Responses Responses are the outputs you will judge: net return, Sharpe, drawdown, turnover, capacity, tail loss, hit rate, exposure concentration, and live slippage. Pick the primary response before testing so you do not promote whichever metric happened to look best. Uncontrolled conditions Markets are not laboratory benches. Regimes, liquidity, volatility, corporate actions, and data revisions move around…
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in