Skip to content
Bayesian modelling of shot generation and conversion in soccer
Branch: develop
Clone or download
Latest commit 90c4aa3 Jan 30, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
doc/literature Rewrite/simplify README Oct 23, 2018
.gitignore Gitignore R cruft part deux Aug 2, 2018
LICENSE Release under LGPLv3 Jan 30, 2019
README.md
analysis.py
requirements.txt Initial commit Aug 1, 2018
run.py
shotconv.stan Add dependence on time elapsed (closes #3) Oct 23, 2018
shotgen.stan Add dependence on time elapsed (closes #3) Oct 23, 2018

README.md

shot-generation

Bayesian modelling of shot generation and conversion in soccer.

Model

The model is described briefly in a blogpost, for full definition consult the shotgen.stan and shotconv.stan files.

Execution

Currently the model is best executed and visualised via IPython, like so:

%run run.py
gen_samples, team_map = run("shotgen.stan")
conv_samples, _ = run("shotconv.stan")

%run analysis.py
team_profiles(gen_samples, conv_samples, team_map)
score_bars(gen_samples)
score_bars(conv_samples)

which produces figures like in the blogpost.

You can’t perform that action at this time.