How to Cluster Basketball Players with K‑Means and Visualize Results in Python

This article revisits a K‑Means clustering case on basketball player data and introduces Matplotlib plotting optimizations.

1. Case Implementation

The dataset contains per‑minute assists, per‑minute points, height, playing time, and age for 20 players. The goal is to classify players into positions (point guard, forward, center) using K‑Means with three clusters.

Full code (shown in images) loads the data, fits sklearn.cluster.KMeans with n_clusters=3, predicts cluster labels, and visualizes the results with matplotlib.pyplot.

The resulting scatter plot shows three distinct groups: red (high scoring), blue (average), and green (high assists, low scoring – point guard).

2. Matplotlib Plotting Optimization

Discusses improvements such as reading data from a file into a matrix, customizing colors and markers, and adding a legend. An example dataset with 96 athletes is shown, and code (image) demonstrates extracting specific columns and plotting.

3. Common Spyder Issues

Lists typical problems when using the Spyder IDE and provides solutions: restoring the editor pane, installing missing packages (e.g., lda, selenium) via pip, fixing a missing console, reinstalling Spyder, and enabling Chinese characters in Matplotlib.