socceraction

Convert existing soccer event stream data to SPADL and value player actions

Showing:

Popularity

Downloads/wk

0

GitHub Stars

307

Maintenance

Last Commit

1d ago

Contributors

15

Package

Dependencies

9

License

MIT

Categories

Readme

Convert soccer event stream data to the SPADL format
and value on-the-ball player actions



pypi Python: 3.6+ Downloads Build Status Code coverage License: MIT Code style: black

This is a Python package for objectively quantifying the impact of the individual actions performed by soccer players using event stream data. It contains the following components:

  • Converters for event stream data to the SPADL and atomic-SPADL formats, which are unified and expressive languages for on-the-ball player actions. Read more »
  • An implementation of the VAEP and Atomic-VAEP frameworks to value actions on their expected impact on the score line. Read more »
  • An implementation of the xT framework to value ball-progressing actions using a possession-based Markov model. Read more »

Installation / Getting started

The recommended way to install socceraction is to simply use pip:

$ pip install socceraction

socceraction officially supports Python 3.7--3.9.

The folder public-notebooks provides a demo of the full pipeline from raw StatsBomb data to action values and player ratings. More detailed installation/usage instructions can be found in the documentation.

Research

For more information about SPADL and VAEP, read our SIGKDD paper "Actions Speak Louder Than Goals: Valuing Player Actions in Soccer" available on ACM (https://dl.acm.org/citation.cfm?doid=3292500.3330758) and Arxiv (https://arxiv.org/abs/1802.07127).

For more information about xT, read Karun Singh's blog post: https://karun.in/blog/expected-threat.html

If you make use of this package or the ideas in our paper, please use the following citation:

@inproceedings{Decroos2019actions,
 author = {Decroos, Tom and Bransen, Lotte and Van Haaren, Jan and Davis, Jesse},
 title = {Actions Speak Louder Than Goals: Valuing Player Actions in Soccer},
 booktitle = {Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
 series = {KDD '19},
 year = {2019},
 isbn = {978-1-4503-6201-6},
 location = {Anchorage, AK, USA},
 pages = {1851--1861},
 numpages = {11},
 url = {http://doi.acm.org/10.1145/3292500.3330758},
 doi = {10.1145/3292500.3330758},
 acmid = {3330758},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {event stream data, probabilistic classification, soccer match data, sports analytics, valuing actions},
} 

Rate & Review

Great Documentation0
Easy to Use0
Performant0
Highly Customizable0
Bleeding Edge0
Responsive Maintainers0
Poor Documentation0
Hard to Use0
Slow0
Buggy0
Abandoned0
Unwelcoming Community0
100