Data Project 9: Which Electric car should you choose?

Machine learning analysis of 478 electric vehicles shows the truth about range, efficiency, and which brands actually deliver.

The electric car market includes too many different models, but here’s what nobody tells you: choosing the wrong electric car could cost you $10,000+ in inefficiency, range anxiety, and depreciation over 5 years.

This isn’t another opinion piece. We analyzed 478 electric vehicles across 22 specifications using machine learning, statistical testing, and clustering algorithms to answer the questions that actually matter:

• Which factors determine range?

• Do premium brands justify their price?

• Is faster acceleration worth the efficiency penalty?

By the end of this analysis, you’ll know exactly which EV segment, brand, and feature set matches your needs, backed by data, not marketing.

Here’s what we’ll uncover using 7 data science techniques applied to real-world EV specifications:

1. Which factors affect electric vehicle range?
   → Technique: Multiple Linear Regression & Correlation Analysis

2. How do EVs cluster by performance and efficiency?
   → Technique: K-Means Clustering & PCA Visualization

3. Does faster acceleration reduce range?
   → Technique: Correlation Analysis & Comparative Statistics

4. Which vehicle segment is most efficient?
   → Technique: ANOVA (Analysis of Variance)

5. Which cars charge fastest relative to battery size?
   → Technique: Feature Engineering & Ranking Analysis

6. Do larger vehicles have worse efficiency?
   → Technique: Correlation Analysis & Volume Calculation

7. Which brands offer the best range and efficiency?
   → Technique: Aggregation Analysis & Brand Positioning

At the end, you’ll learn these techniques, how to apply them, and the results will help you make information-based decisions.

---

There are four related files in the gDrive.

• Data

• Jupyter Notebook includes code

• Jupyter Notebook turned into a PDF with code

• Report created with AI

---

What do we offer to the paid subscribers?

Paid subscribers get more than articles; projects with code, custom GPTs, and prompts designed after thousands of hours of experience.

This week’s project: Data Project 8: I Analyzed 182,111 Chicago Crimes. The Arrest Lottery Has Patterns You Won’t Expect

This week’s topic: Stop Googling for Hours. Use These Research Prompts Instead.

And last week, we tuned in with AI news and discussed this AI shift,

• What happens when governments start regulating not how AI works, but how it talks?

Join the community to build real projects, master new prompt techniques (future prediction, simulation), and stay ahead of critical shifts before they reshape your industry 👉