Credit to Dataquest Welcome to the NBA Game Predictor repository, where data scraping meets predictive analytics for NBA enthusiasts and data science aficionados alike! This project leverages Python to scrape real-time NBA statistics from Basketball Reference, harnessing the power of data to predict game outcomes.
The goal of this project was to see if I could make use of data scraping to build an effective model for NBA game prediction. Though not currently able to be used on daily NBA games, this model is effective on previous NBA seasons, correctly guessing games at a 63% clip. However, if one did update the data with current daily games, there is a way for the model to be used as such.
The code is a great baseline for looking at what important factors determine basketball games. Using these 'predictors', one could use a ridge classifier with those predictors to determine the outcomes of NBA games.