Underfitting happens when a statistical model fails to sufficiently capture the inherent patterns and relationships within the data. As a result, the model's performance on both training and new, unseen data remains subpar. It often presents itself as overly simplistic, with the model lacking the complexity needed to accurately represent the data's true nature. For example, using a simple linear model to fit data that follows a complex, nonlinear distribution would likely lead to underfitting, as the model cannot adapt to the data's intricate features.
