The raw material for statistical learning are data. Different applications of statistical learning can involve very different data: stock prices in finance, amplitudes of birdsongs in biology, or neural activity levels in neuroscience. But rather than individual schemes for specific applications, statistical learning provides general tools for data analysis: for example, the same methods might indicate that the stock market became more volatile after the last financial crisis and that the corona pandemic influenced birdsongs—even though companies and birds have little in common. This versatility roots in the fact that statistical learning formalizes data in an abstract mathematical framework called probability theory. In this part of the book, we introduce this mathematical framework, and we use it to describe given data.