The Impact of Champion Pick Rate on Win Rate in League of Legends
With the release of Patch 5.4, discussions on champion balance in League of Legends have taken center stage on various forums. Much of the debate has revolved around junglers, specifically which champions are strong and whether the top tier junglers are too strong compared to others. The focus has also been on Riot’s balance goals.
In one particular discussion on Reddit, the champion being discussed was Sejuani. One user claimed that Sejuani was useless and unplayable, questioning Riot’s decision to leave her in such a state. However, another user pointed out that Sejuani actually had a 54% win rate in ranked solo queue, making her the 5th best champion in the game. The user argued that Sejuani’s high win rate was due to her lower pick rate, meaning that only experienced Sejuani mains were playing her. They also mentioned that Jarvan, who had a lower win rate, was considered stronger than Sejuani but had a misleading win rate because bad players were influenced by his popularity in professional play. This comment garnered agreement from other users.
League of Legends has a player base of 27 million. Upon analyzing champion data from the past week, which compiled data from over 1 million games, it became evident that popular opinion tends to generalize the 40,000 players who chose Sejuani as experienced masters of the champion, while dismissing the 85,000 players who chose Jarvan as inexperienced newcomers just following the hype.
This generalization seemed laughable to me. Previous research and word of mouth have shown that there are champions at both ends of the spectrum in terms of win rate (Urgot being a notable example). Is there truly a correlation between pick rate and win rate? To find out, I decided to study the numbers.
I collected pick rates and win rates from games played in the past week, totaling over 6 million picks in all.
In the graph below, each dot represents a champion. The further to the right a dot is, the higher the win rate, and the higher it is positioned, the greater the number of picks. Since there are 121 champions, the graph can get messy, so I decided to sort the champions.
Based on win rate, any champion with a win rate of 50% or higher is categorized as having a “high win rate,” while those below 50% are considered to have a “low win rate.”
As for popularity, the average pick number for each champion would be 50,000. Champions who are picked more often than that are classified as popular, while those with below average pick rates are labeled as unpopular.
High win rate champions are almost equally split between popular and unpopular picks. On the other hand, champions with poor win rates are predominantly unpopular picks. This data confidently refutes the notion that a champion with a low pick rate is more likely to have a high win rate. In fact, a high pick rate champion is much more likely to have a high win rate than an unpopular champion.
I believe that players are becoming more informed about the game thanks to the abundance of tools for checking champion stats, finding guides, the growing viewership of League of Legends esports, and the popularity of online communities like Reddit. As a result, more players are gravitating towards the stronger champions.
The top 10 most popular champions have an average win rate of 51%, with 7 out of the 10 having win rates above 50%. Not only are these champions strong, but they also have well-designed skillsets, interesting backstories, and captivating themes that make them enjoyable to play.
Among these popular champions, there are many formidable contenders who can make a significant impact in games.
I also noticed a trend among champions who are popular but have a below average win rate. These champions often have a high skill cap, making them more challenging to play effectively. Examples include Thresh, Lee Sin, Riven, Vayne, and Zed.
The 10 least picked champions have an average win rate of 46.6%, with only one of them (Poppy) having a win rate above 50%. It can be concluded that these champions are currently in a poor state.
Closing Thoughts
Contrary to the myth that low pick rate champions have a higher win rate, the data shows that for every low pick rate champion performing well, there are two performing poorly. It is important to recognize that pick rate does not guarantee a high win rate.