Collaborative project to create a number of test models for gambling in League of Legends
- Have built structure to request data from pro matches, given either the game
- Can also get all games in a tournament by chaining the tournament lookup and series lookup
- To see data in its current form, look at (and run) doc/examples/proof_of_concept
- Also need to make the output data readable and output into some csvs (CURRENT WORKING TASK)
- Need to automate collation of data into csv's given a tournament (focus on LCK to begin with)
- Perhaps also make a script to request game data once a week automatically?
- Store data locally on server
- Interface with riot API to pull data about amateur games
- Create summary statistics and store those (likely too much data otherwise)
- Further post-processing of data
- Develop tools to use data to have more useful predictors
- Predict outcomes of games in advance
- Predict other stats about games in advance