CFB data is an open-source repository of FBS college football data. It is a combination of original data from contributors and automatic extraction from sites like Wikipedia.
Data is provided in CSV and JSON format and is largely denormalized for ease of use.
College football is one of the great strategy sports of the world, producing an immense amount of data, but there is no simple-to-use and open repository for it to reside.
Here is a list of the provided datasets and information on using them.
- Teams - a CSV, sourced from Wikipedia, of all of the teams, their locations, and mascots.
- Stadiums - a CSV, sourced from Wikipedia, of all of the stadiums, their locations, and capacities.
- Coach Tenures - a CSV, sourced from Wikipedia, of coaches and their respective tenures at schools.
CFB Data is in development and there will be likely be sweeping changes over time. However, my (@coffenbacher) general approach is to work on three tiers of projects:
-
Data persistence - The open repository https://github.com/coffenbacher/cfb-data will serve as the repository for acquired data. Github allows the data to be discoverable, natively versioned, and CDN hosted for hotlinking using http://rawgit.com/.
-
Data acquisition - Multiple open- and closed-source projects by both myself and others will automatically contribute data to the the cfb-data repository. Data can be either committed directly by approved projects or provided via pull request.
-
Data client - Client libraries in various popular languages to access the data and make usage clean.
-
Data visualization - Closed-source visualization projects, like blogs and websites, are welcome to use the data. Of course, open-source projects are welcome too, but I expect most users at this tier will be doing closed analyses.
This is very much a work in progress - please contact me at charles.offenbacher@gmail.com to discuss collaborating. I'd love to have help!