I have attached included a link to matches.json
matches.json
(the first try didn't work because GitHub Gist doesn't support large files) which is a weighted dataset that I randomly generated with specific weight values.
The format of each match is like so:
[
// Team 1
[...], // Champion IDs
// Team 2
[...], // Champion IDs
// 0 means team 1 won, 1 means team 2 won
0 || 1
]
Find the p-value (calculated probability) for each champion winning a match.
Find the p-value of a champion winning a match, given the enemy team's composition (conditional probability).
The challenge will be graded on how closely the results reflect the weighted probabilities. There are some clear statistical anomalies in the data set. Identify which champions fall outside of a standard deviation.