League of Legends - Buffing and Nerfing

outlier detection
five number summary
z-scores
Game play statistics for League of Legends champions in two different patches. Can detecting outliers indicate champions that should be enhanced (buffed) or scaled back (nerfed)?
Author

George Charalambous, Ivan Ramler, and A.J. Dykstra

Published

July 1, 2023

Motivation

League of Legends is a 5 v. 5 multiplayer online battle arena (MOBA) game developed by Riot Games. They collect data to evaluate the effect of every champion, adjusting and fine-tuning various aspects associated with each champion, to ensure fair and competitive gameplay. Through updates (patches) they can reduce the power of champions that are too strong (known as nerfing) or enhance the abilities of less successful champions (nerfing).

Data

Each row is one of the 162 League of Legends champions. The two datasets show play results (e.g. win rates) for two consecutive patches of the game (12.22 and 12.23).

LOL_patch_12.22.csv, LOL_patch_12.23.csv
Variable Description
Name Name of the champion
Role Role of the champion in a game
KDA Average kills, deaths and assists associated with each champion
WRate Win rates of each champion
PickRate SPick rates of each champion
RolePerc Percent of time the champion is used in the expected role
BanPerc Ban percentages associated with each champion

Questions

  1. Look for outliers (e.g. in win rate) for the 12.22 patch to identify champions that might need to be buffed or nerfed.

  2. Look at results for the 12.23 patch to see if outliers identified from 12.22 might have been adjusted of if any different champions are now outliers.

References

Lol champion stats, 12.22 master, win rates. METAsrc. (n.d.). https://www.metasrc.com/5v5/12.22/stats?ranks=master

Lol champion stats, 12.23 master, win rates. METAsrc. (n.d.-b). https://www.metasrc.com/5v5/12.23/stats?ranks=master