Skip to content
A machine learning model that predicts the statistics of NBA's 2019-2020 season rookies
Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
College and Rookie Trends.ipynb
College&Rookie Data Scrapper.ipynb
Fixed CollegeRookieStatLog2.ipynb
Full College Clustering.ipynb
Prospect Scraper (2019).ipynb
Rookie Predictions .ipynb

NBA Rookie Statistics



A mathematical machine learning model designed to predict the regular season averaging statistics for incoming rookies, by applying the trend of all the NBA player’s (1980 - 2018) college careers compared to their first season in the league, to the 2019 rookies college career statistics. I built this model using Python, and I’m currently working on a UI to display the calculated data.

Data Preparation

Before actually playing with and analyzing the data, I used BeautifulSoup to scrape all the required data, which included the following:

  • Year-by-year in-game NBA statistics of every player from 1980 to 2019
  • Year-by-year in-game College statistics of the every player accounted for above
  • In-game College statistic of every in-coming NBA rookie for 2019


Following are the concepts that were applied to predict the 2019 in-game statistics of incoming rookies:

  • Correlation Matrix
  • Linear Regression
  • Extra Trees
  • Random Forest
  • XGBoost
  • Feed Forward Neural Network
  • Differences


The following are a few examples of the predicted statistics for the 2019-2020 Rookies:

Players Points Assists Rebounds Steals Blocks Draft Pick
Zion Williamson 13.0 2.4 6.2 1.1 0.8 #1
Darius Garland 8.1 1.9 2.9 0.6 0.2 #5
Coby White 7.8 2.3 2.2 0.6 0.2 #7
Cam Reddish 6.1 1.1 1.6 0.6 0.2 #10
Bol Bol 9.4 1.1 4.9 0.5 0.8 #44


Data: BeautifulSoup, Pandas

Visualization: SeaBorn, Matplotlib

Analysis: TPOT, XGBoost, SkLearn

You can’t perform that action at this time.