Code for posts on thespread.us.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
notebooks
qb_forecasting
ranking
2013_nfl_pbp_data_through_wk_12.csv
2013lines.sql
README.md
clustering_wrs.py
load_pbp_data.py
nfl2013lines.csv
setup.sh
team_name_map.csv
test_data.sql
training_data.sql

README.md

thespread

This is the repository for posts on thespread.us. In order to reproduce the posts on the spread, you will need to purchase play-by-play data from Armchair Analysis for $25.

DISCLAIMER: I do not receieve any compensation if you buy the play-by-play data. It's just the best, cheapeast, and most comprehensive play-by-play-data available.

Dependencies

To use the files in this repo, you should be on a UNIX-based system (Mac OS X should work, though these files have only been tested in Linux Mint 15 and Ubuntu 13.10). You will also need the following dependencies installed:

You can clone this repo where you would like to install these files and use setup.sh to get all of the databases created. The syntax is:

setup.sh [name of Armchair Analysis zip file] [desired name of database]

Python library dependencies

I use a number of Python libraries. If you have done any kind of data science or statistical work in Python before, you probably have them installed. You should be able to install them with pip or setuptools. Some of these libraries include:

  • pandas
  • scikit-learn
  • numpy
  • matplotlib
  • seaborn

If you have any questions, please let me know. Unfortunately I cannot offer tech support, but I'm glad to help where I can.

Trey