NOTE

Dear reader,

this page is outdated and mainly composes of work from outside my studies. Since I have been compelting a masters degree in data science, I have had little time to update this page with some of my newer/better work. Feel free to look, but if you would like to see more evidence of proficiency, please do get in touch: mattbenyon@hotmail.com

Matt Benyon

This page is a summary of some projects that I have undertaken to demonstrate by ability as a data analyst and data scientist. I hope you take the time to read through the summaries/write-ups and/or view my code avaialable in the GitHub repository.

1. Basketball four factors analysis

Linear Regression of the basketball ‘four factors’ from previous seasons to predict a teams win totals in the current season.

Project completed, write-up is in progress

2. NBA 2K player ratings analysis

• Acquiring the data: Scraped data from 2kratings.com, basketballreference.com and downloaded .csv file from FiveThirtyEight.
• Cleaning the data: Cleaned the data by renaming columns, dropping nan’s, splitting strings so player names matched format across dataframes etc.
• Plotting the data: Used MatPlotLib to create plots of All-Star votes against NBA 2K ratings, FiveThirtyEights RAPTOR WAR metric against 2K ratings among others. Used Scipy package to fit a line to the appropriate figures. Provided a logarithmic version to show correlation.
• Project was completed at own leisure as a test of my developing scraping, cleaning and analysing skills. I wanted to test a hypothesis I had that NBA 2K player ratings were swayed more by popularity than by defensive impact. I verified this hypothesis but also acknowldged that a players popularity is mostly a product of their offensive ability. Overall I found that defensive impact had little correlation with a players rating and player popularity correlated strongly.

An example figure from the study:

Feel free to check out my write up available at the hyperlink on the heading or through my GitHub page.

Write-up coming soon once completed undergraduate exams