CNU Men's Basketball EDA and ML Project
Background
I have completed the Jovian Data Analysis and Machine Learning courses where I performed analysis and trained ML models from a varierty of open source datasets. And for both courses there were capstone projects I planned to use a dataset that I had more intimate knowledge about, my college basketball team's historical statistics.
However, in my naivete I was assuming that the availability and cleanliness of data in the "wild" would be similar to the datasets that I was presented on Kaggle. I quickly learned that was indeed NOT the case. Even for my universitiy where the stats have been religiously and dilligently taken for over a decade now, getting the data was not nearly as easy of a task as I initially imagined. Since then, I have been able to gather and clean the data myself, and use it in other applications like my CNU Men's Basketball Data Dashboard and eventually the data will be available using an API I am developing.
But in the mean time, I have the data in a format that I can work with (CSV) where I can perform some analysis on.
Overview
As a former CNU Men's basketball player and extreme data nerd, it only felt right to use my newly learned skills of data analysis and machine learning to enjoy the intersection of those two hobbies.
Plan is to take the data that I have gathered from the 12 years of CNU and Opponent team data to discover what determines whether or not CNU wins a basketball game. The goal of this project is to use exploratory data analysis to understand the data that was collected about the team, discover which are most related to winning, and then hopefully use that insight to develop a classification machine learning model to predict/classify if CNU will win a game or not.