This is the fourth blog post in our chess series. We will take a look at how machine learning models detect increases in the use of popular chess openings as anomalies.