Skip to content

smitsam555/sportsipy

 
 

Repository files navigation

Sportsipy: A free sports API written for python

Started in early 2023, this project is an active fork of roclark/sportsipy, fixing basic issues related to site layout changes, python dependencies, and failing test cases. Note that not all existing functions work due to changes on the source site, but fixes will be integrated as they are identified.

Sportsipy is a free python API that pulls the stats from www.sports-reference.com and allows them to be easily be used in python-based applications, especially ones involving data analytics and machine learning.

Sportsipy exposes a plethora of sports information from major sports leagues in North America, such as the MLB, NBA, College Football and Basketball, NFL, and NHL. Sportsipy also now supports Professional Football (or Soccer) for thousands of teams from leagues around the world. Every sport has its own set of valid API queries ranging from the list of teams in a league, to the date and time of a game, to the total number of wins a team has secured during the season, and many, many more metrics that paint a more detailed picture of how a team has performed during a game or throughout a season.

Important: This fork is not the reference on PyPI. To include this fork in your requirements:

pip install git+https://github.com/davidjkrause/sportsipy@master

If the bleeding-edge source version of sportsipy is desired, clone this repository using git and install all of the package requirements with PIP:

git clone https://github.com/davidjkrause/sportsipy
cd sportsipy
pip install -r requirements.txt

Once complete, create a Python wheel for your default version of Python by running the following command:

python setup.py sdist bdist_wheel

This will create a .whl file in the dist directory which can be installed with the following command:

pip install dist/*.whl

The following are a few examples showcasing how easy it can be to collect an abundance of metrics and information from all of the tracked leagues. The examples below are only a miniscule subset of the total number of statistics that can be pulled using sportsipy. Visit the documentation on Read The Docs for a complete list of all information exposed by the API.

from sportsipy.nhl.teams import Teams

teams = Teams(2018)
from sportsipy.nba.teams import Teams

teams = Teams()
for team in teams:
    print(team.name, team.abbreviation)
from sportsipy.nfl.teams import Teams

teams = Teams()
lions = teams('DET')
from sportsipy.ncaab.schedule import Schedule

purdue_schedule = Schedule('purdue')
for game in purdue_schedule:
    print(game.date)
from sportsipy.ncaaf.boxscore import Boxscore

championship_game = Boxscore('2018-01-08-georgia')
print(championship_game.away_interceptions)
from sportsipy.mlb.boxscore import Boxscore

game = Boxscore('BOS201806070')
df = game.dataframe
from sportsipy.fb.team import Team

tottenham = Team('Tottenham Hotspur')
print(tottenham.goals_scored)

Two blog posts detailing the creation and basic usage of sportsipy can be found on The Medium at the following links:

The second post in particular is a great guide for getting started with sportsipy and is highly recommended for anyone who is new to the package.

Complete documentation is hosted on readthedocs.org. Refer to the documentation for a full list of all metrics and information exposed by sportsipy. The documentation is auto-generated using Sphinx based on the docstrings in the sportsipy package.

Sportsipy contains a testing suite which aims to test all major portions of code for proper functionality. To run the test suite against your environment, ensure all of the requirements are installed by running:

pip install -r requirements.txt

Next, start the tests by running py.test while optionally including coverage flags which identify the amount of production code covered by the testing framework:

py.test --cov=sportsipy --cov-report term-missing tests/

If the tests were successful, it will return a green line will show a message at the end of the output similar to the following:

======================= 380 passed in 245.56 seconds =======================

If a test failed, it will show the number of failed and what went wrong within the test output. If that's the case, ensure you have the latest version of code and are in a supported environment. Otherwise, create an issue on GitHub to attempt to get the issue resolved.

About

A free sports API written for python

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 96.1%
  • HTML 3.9%