Overfitting
Also called: curve fitting, data mining, in-sample overfit
Overfitting is building a model that captures the noise in your sample rather than the signal, so it performs beautifully in-sample and fails on new data. It's the central hazard of quantitative finance, where thousands of strategies can be tested against the same price history until one looks brilliant by chance. The defences are out-of-sample testing, cross-validation, and being honest about how many variants were tried.
Want more than a definition? Learn it in Regression Range →