Why AFL? You hate sports
In my team at work we do an annual thing where we play Fantasy AFL. While I have no love for this pass time or sports in general, I do think it would be a fantastic project for my own mini data warehouse.
There are quite a few things I could bring in: match stats, players, teams, home/away, game day temperatures and weather. Hell, with a little bit of work I could bring in stuff like news reports about players misbehaving and see how that affects them.
A cool thing to do would be perform cohort analysis on the players themselves. For instance, in the range of 400,000 players, how many went up at the end of each match, or down.
Data
The first thing we need is good data, and luckily Kaggle has my back. There is a dataset that has information from 2012 through to 2023. I think this is a large enough dataset to handle my needs.
The only thing it doesn’t appear to have is player pricing over time. However, I have found a few articles that will hopefully let me reverse engineer the pricing based on the information I have.
Code
The code can be found on my repo.