Analyzing NBA Data in Python | NBA Data Analytics Project (part 2/2)

Link to this Jupyter Notebook:

Contact me:

Add to your sports data analytics project portfolio with part 2 of this video series. I break down and visualize three main topics regarding NBA stats:

1) Player stat correlations
2) Player minutes and scoring distributions (Playoffs vs. Regular Season)
3) How the game has changed over the past decade (Playoffs vs. Regular Season)

Leave a comment with any other topics you'd like to see me cover!

Intro: 0:00
Data cleaning & preparation: 3:27
Stat correlations (Section 1/3): 9:34
Distributions of minutes played (Section 2/3): 21:03
How the game has changed (Section 3/3): 34:29
Comparing Regular Season vs. Playoffs: 56:00

[jameskelly]

Thank you, Alex! I have tried several other videos to get this kind of stuff working, but all seemed to have a feature that is no longer working. Your stuff is working, so I can use what you taught us about, and I can now use this for what I am working on.

[richardjozsefkuba]

Awesome video, I was lazy to study plotly in this detail so I saw lot of good tricks. Thanks ...data analytics and NBA two things I really like :)

[futureverse]

Thank you as a former athelte who is trying to learn data science, I am impressed nice work!

[GreyBulll]

Great Tutorial a lot for college project as well as self learning.
Although Indians do not follow NBA much but after this project i'll surely follow. Keep posting tutorials. PEACE

[allenc]

Thanks for the video Alex! Im a fan of basketball myself. It is super interesting and exciting to do such data analytic project with something I love as my first ever project!

[pdmarino]

Thanks for this. I'm just starting out in Python so these kinds of studies are useful to learn. Do you have a coy of the Jupyter notebook code anywhere? Was following the two vids but wasn't easy to read everything off the screen. Thx.

[Tannerbrodess]

Could you do a tutorial of doing college football data using the import cfbd?

[nickhi]

can you explain why we converted it to data per min i am struggling to understand why we didnt just leave it

[richardjozsefkuba]

hint: when you double click the data series in the legend plotly will hide all the others

[mawkuri]

if this can predict over under and spread. this would be awesome🤣

[lplp]

This a great tutorial video. Could you do it one like this for soccer too like working with premier league data?