League of Legends

Is Renekton OP? The Winrate Paradox

LeagueofLegends7 - Is Renekton OP? The Winrate Paradox
Loading...

In the last few weeks, there has been a lot of discussion around Renektons strength in pro play. One of the main arguments for Renekton has been his high winrate in the LPL, LCK and LEC.

(Image from @ lplenglish)

This seems to indicate that Renekton is a good pick, right? Not necessarily, because context matters. If a champion for example only works as a counterpick it probably wouldn't be a great Idea to blind pick this champion, even though he has a high winrate. Other factors play a big role too, like teamcombs, Player proficiency on a champion, samplesize and so on.One thing that often gets overlooked is that high winrate teams who play a champion a lot (like V5 or TES in the LPL) can inflate his winrate. To explain this a little bit better, here is a little example from an imaginary League with 3 teams:

https://preview.redd.it/vhnpzhw5wbm51.png?width=612&format=png&auto=webp&s=a193cfe51c97da132cb03b221bd8249007bd8372

As you can see, his winrate in this League is 53%, over 70 games. What you can also see is that Team 1, who plays Renekton more that T2&T2, has a pretty high winrate on the champion. Now take into account the winrate of these 3 teams over all games in their split:

https://preview.redd.it/sii9r2r9wbm51.png?width=398&format=png&auto=webp&s=c1202b73172e6e096ee087232d09c3a98e73bee9

Every single team has a lower winrate on the champion than in their overall split! And the champion still has a 53% winrate, even though every single team is more likely to lose when they play him! Weird, right? This is because Team 1 for some reason plays the champion a lot, and thus "inflates" his winrate.

Загрузка...

A better Stat than "Winrate"?

An easy way to make a compact Stat is by adding the Values on the right and dividing them by the number of values, so we have an "average Renekton multiplier". Lets do this for the LEC:

https://preview.redd.it/fg6ibu9zzbm51.png?width=538&format=png&auto=webp&s=4e35576e1bdd8abeadffa0cd180242627a90944a

Here, the "average Renekton multiplier" is 0.7709, so you are 23% less likely to win with Renekton than with a random champion. The problem with this is that we don't account for the differing number of games the teams played on him: The 0% WR from 1 VIT game counts as much as the 100% WR from 5 MAD games, that's why I think we should give more weight to teams who play the champion more:With this adjustment, the "weighted Renekton mulitiplier" is 1.13, so you are more 13% likely to win on Renekton than on on a random Toplaner. There is another problem with this though: samplesize. The LEC only played 16 games of Renekton this split. Lets look at the real Leagues: LPL and LCK.

In the LPL, Renekton has a Winrate of 52,34% over 128 games. If you take into account that high winrate teams seem to pick the champion more than others, this drops to 48.42%!

For the LCK, with 44 games and a 57% winrate, it drops even lower, to 47.44%.

These Stats are not perfect, because it can't take into account other context, but it's still are more accurate singular value for a champion than "winrate". The goal of this is not to debate if Renekton is good or not, but to show how adding a bit of context changes these very oversimplified metrics drastically.Do you agree? Did I do something wrong? Interested in hearing your opinions.

TLDR: Statistics are not always as they seem, so handle with care!

Source: Original link


Loading...
© Post "Is Renekton OP? The Winrate Paradox" for game League of Legends.


Top 10 Most Anticipated Video Games of 2020

2020 will have something to satisfy classic and modern gamers alike. To be eligible for the list, the game must be confirmed for 2020, or there should be good reason to expect its release in that year. Therefore, upcoming games with a mere announcement and no discernible release date will not be included.

Top 15 NEW Games of 2020 [FIRST HALF]

2020 has a ton to look forward to...in the video gaming world. Here are fifteen games we're looking forward to in the first half of 2020.

You Might Also Like

Leave a Reply

Your email address will not be published. Required fields are marked *