Master Principal Components Analysis in under 20 Minutes (ft. Pete Crow-Armstrong’s Stats)

What do you get when you combine baseball stats with one of the most powerful techniques in data science?

In this video, we break down Principal Components Analysis (PCA) using game-by-game data from Pete Crow-Armstrong — one of the most exciting young players in the game.

We walk through:

What PCA actually does

Why we use it (vs. factor analysis)

How to interpret communalities, loadings, eigenvalues, and rotations

And how PCA reveals multiple "styles of play" in Pete’s game

Whether you're a stat student, a data science learner, or just a baseball nerd, this is PCA like you've never seen it before.

⏱️ All in under 20 minutes. Let’s go.

🎯 Download the FREE Method Map PDF: 👉 https://curious-modeling.kit.com/methodmap
💾 Data: Game-by-game stats from Pete Crow-Armstrong (2023–2024 season)
🎓 Tool: Principal Components Analysis with Varimax rotation in SPSS

📈 Subscribe for more modeling demos, tutorials, and sports + stats breakdowns.

0:00 Intro and the Data
1:15 Why Principal Components Analysis?
1:28 Running the Model
2:38 Why PCA and not Factor Analysis?
6:15 Model Summary Walkthrough
3:38 Rotation Methods
4:13 Communalities Rule of Thumb
5:59 Variable Extraction Interpretation
7:37 Total Variance Explained Table
9:00 Scree Plot
9:54 Component Tables
11:57 Loadings Interpretation – the PCA Model
14:46 Component Transformation Matrix
15:54 Conclusion

Видео Master Principal Components Analysis in under 20 Minutes (ft. Pete Crow-Armstrong’s Stats) канала Curious Modeling