Soccer Analytics with Machine Learning

Chapter 5. Predicting Soccer Outcomes with Classification

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 5th chapter of the final book. Please note that the GitHub repo will be made active later on.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

In Chapter 4, we modeled numerical outcomes such as market value and goals scored. Many soccer questions, however, are not about magnitude but about category: whether a shot becomes a goal, whether a pass is completed, or which result a match produces. That is the setting for classification.

Classification keeps the same overall workflow as regression: define a target, choose features, train a model, and evaluate it carefully. What changes is the task itself. Instead of estimating a continuous value, we assign an observation to one of several discrete classes. Think about questions such as:

• Will this shot be a goal or not?

• Is this pass likely to be successful?

• Will the home team win, lose, or draw?

• What type of pass will a player attempt under pressure?

These are all classification problems. The answer is not a number on a sliding scale, but a label or a probability over labels. That shift may sound small, but it changes both ...