We propose new methods to identify, quantify and describe the grammatical-stylistic changes that take place during the lifetime of an author. To examine the strength of the chronological signal of change, we first developed a method to calculate if a distance matrix of literary works contains a stronger chronological signal than expected by chance. 10 out of 11 corpora showed a higher than chance chronological signal. Second, we proposed a machine learning task: predicting the year in which a work was written. The accuracy and the amount of variance that is explained by the model were high for most authors we studied. After applying a feature selection algorithm, we examined the most important ones, i.e. patterns that have the greatest influence on idiolectal evolution.