When building a model, more information sounds better — but adding every variable you can find often makes predictions worse. This chapter gives you two families of tools to keep your model smart and simple: feature selection that chooses the most helpful variables, and regularization that gently reins in the ones you keep.