Soccer Analytics with Machine Learning

Chapter 8. Feature Engineering for Soccer Analytics

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 8th 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 7, we saw that even highly flexible models still depend on the quality and structure of their inputs. This chapter turns to that representation problem directly.

Exploratory Data Analysis helps us understand what is in the data. Feature engineering is where we decide what the model should actually learn from it.

That distinction matters. Raw soccer data is rich, but it is rarely organized in the form that a model needs most. Match results tell us what happened, but not whether a team is improving. Event data tells us where actions occurred, but not which patterns are most important. A league table tells us who is ahead, but not how strong a team really is once schedule difficulty, venue, form, and shot quality are taken into account.

Feature engineering is the step where analysis becomes modeling. It is where we translate soccer knowledge into structured signals: recent form instead of season totals, venue-specific performance instead ...