soccer-prediction

poisson.py

code for https://steemit.com/python/@stevencurrie/soccer-predictions-using-python-part-1 & https://steemit.com/programming/@stevencurrie/soccer-predictions-using-python-part-2

soccerprediction.py

updated code for https://steemit.com/programming/@stevencurrie/soccer-predictions-using-python-part-3

updated code for https://steemit.com/programming/@stevencurrie/soccer-predictions-using-python-part-4

updated code for https://steemit.com/programming/@stevencurrie/soccer-predictions-using-python-part-5

Successfully tested on standard windows installation of python 3.6.

copied soccerprediction.py to it's own folder.

to install dependencies run

pip install pandas
pip install requests
pip install beautifulsoup4

or pip install -r requirments.txt

then run

python soccerprediction.py -t

This will download data for the English Premier League (default -c "England" -l "Premier League") and run tests on the data to find the best settings for the -y and -b options.

At this time, it returns 400 for HISTORY and 70 for cutoff.

python soccerprediction.py -y 400 -b 70

will run the prediction and printout to the console any games that include a probability higher than the cutoff of 70%.

You can add the -d YYY-MM-DD option to predict a few days in advance. Not recommended to go to far as this would decrease the accuracy.

python soccerprediction.py -y 400 -b 70 -d 2017-09-30

Currently only checking to Home/Draw/Away - If you want to add checks for over/under, both to score etc, feel free. A spreadsheet (.CSV) is saved for each competition in the data folder in the same location as soccerprediction.py