Clustering of NBA Teams Across Multiple Years

The study seeks to provide an objective foundation for NBA team rankings beyond win-loss statistics. The research uses machine learning to examine offensive efficiency, defen- sive efficiency, and win % from the last four NBA seasons utilizing sports analytics and big data. The study uses PCA and K-Means clustering to classify team performance tiers. Teams with balanced offensive and defensive measures typically rank higher in clusters, establishing the technique as a more thorough evaluation tool. Moderate accuracy in error analysis shows the balance between analytical depth and practical applicability. The research found that this data-driven paradigm helps comprehend team relationships and performance, which might be used in strategic decision-making and other sports. Refinement may incorporate time patterns and player-specific data.