likarajo / fpl_ai

Artificial Intelligence guide for Fantasy Premier League

AI Guide for Fantasy Premier League

Artificial Intelligence guide for preparing my team for Fantasy Premier League.

Fantasy Premier League

A.K.A FPL in short, it is a game in which participants assemble an imaginary team of real life footballers for a gameweek in the English Premier League and score points based on those players' actual statistical performance or their perceived contribution on the field of play. It requires predicting and foreseeing the future of the players' performance.

The fpl_AI Bot

• It gives suggestions based on insights on what steps to take to maximize the FPL team points.
• It makes the suggestions for player transfer if needed based on:
  • upcoming fixtures
  • players' form
  • team budget
• It sends the suggestions or advice automatically via email a few hours before the FPL Gameweek deadline.

Workflow

Fetch and organize data

• Scrape data
  • Parse and store all the data

cd src/scraping
python global_scraper.py
python teams_scraper.py ${TEAM_ID}

• Update existing data with new data

cd src
rm -rf raw_data/2019-20/players
mv scraping/data/2019-20/players raw_data/2019-20/ -f
mv scraping/data/2019-20/players_raw.csv raw_data/2019-20/ -f
mv scraping/data/2019-20/fixtures.csv raw_data/ -f
mv scraping/data/team_data/my_team.json raw_data/ -f
mv scraping/data/gameweeks.json raw_data/ -f
rm -rf scraping/data

• Clean and preprocess the data

python data_cleaner.py
python data_maker.py

• Run the bot

cd src
python ai.py

• Notify the user

cd src
python ai.py

Links

• Automating GitHub Workflow
• GitHub scheduled events

Data

Kudos to vaastav for the data

Concept

Kudos to ravgeetdhillon for the concept