Overfitting(Overfitting / curve fitting)
Fitting a strategy's parameters so closely to past data that it fails to work in the future; the biggest pitfall of backtesting.
Also written as: aşırı uyum
Signs: a large number of parameters, results degrading sharply with small changes, perfect results on a single market/period. A strategy with few rules that performs reasonably across different markets and periods is better than one that is perfect in a single place.
Safeguards: out-of-sample testing, walk-forward analysis, parameter robustness scanning.
Related terms
Where it appears in the guides
Guides whose text uses this term; most frequent first.