Backtesting
A rigorous simulation framework that evaluates the historical performance of a predictive model or trading strategy using strictly past observational data.
Think of It Like This
Like reading yesterday's newspaper to see if the stock picks you would have made actually ended up making any money in reality.
By feeding the model data chronologically as if it were running in real-time, practitioners can assess how it would have performed before deploying it live. The biggest pitfall in backtesting is data leakage, where future information accidentally bleeds into the training set, causing unrealistically high performance metrics that immediately collapse in production.