Welcome to the Forum Archive!

Years of conversation fill a ton of digital pages, and we've kept all of it accessible to browse or copy over. Whether you're looking for reveal articles for older champions, or the first time that Rammus rolled into an "OK" thread, or anything in between, you can find it here. When you're finished, check out the boards to join in the latest League of Legends discussions.

GO TO BOARDS


Champion Usage at MLG Anaheim (a closer look)

12
Comment below rating threshold, click here to show it.

Krebstar

Senior Member

06-11-2012

Hey all,

I know there's a lot of stats flying around with calls for nerfs today, so I just wanted to throw my 2 cents in about how these stats should be considered, using stats I was able to find at https://elobuff.com/events/2012-mlg-anaheim/champions. If you're in a hurry, you can skip to the spreadsheet for the results, but I'll explain myself a bit here.

What's important in determining balance within a class isn't how many games they were picked/banned, but how many games they were picked/banned compared to the case where all members of that class were picked equally. Mind you, this doesn't account for higher-level things like the fact that some champs have more niche roles than others, but it's a good first approach. In this way, it's much less surprising for a specific support to be used in 50 games than it is for an AD ranged carry, because the support class only has ~8 champs (Lulu, Nunu, Soraka, Sona, Janna, Taric, Alistar, Leona). Therefore, in the attached spreadsheet, I have made assumptions about the size of each class, calculated the number of games each would have been played if they were equally used ("Expected" column), then calculated the % of games this champion was over or underplayed ("%Over" column). If you believe popularity is a good metric for balancing a class, I will argue that the %Over results are what should be considered, NOT raw number of games played.

Additional Notes:
I searched by class on elobuff, but they like to divide classes not by lane, so I made some corrections (e.g. Kayle is listed as ADC, Nautilus as support, Urgot as bruiser)

I have left unorthodox picks out of consideration (e.g. not counting Blitz or Karma as supports)

I haven't calculated anything for top lane because I couldn't get meaningful stats for "bruisers" from elobuff. I also don't like my APC numbers because champions like Vlad, Rumble, and Kennen are counted there, despite probably laning top. Jungle is messy, but I think the number of champions is probably about right, and therefore the results.

The calculation for "%Over" is: (#games played - #expected)/(#total games)*100. So for Janna, the support class has 8 champs played over 172 games, expected = 172/8 ~21.5, %Over = (52-21.5)/172*100 ~ 17.7%.

Conclusions:
Each of the classes have a few highly overused champs. APC looks more balanced, but I don't like these numbers as much because they likely include a lot of the AP top lanes, which I feel should probably be compared against other top laners. I've also underestimated the size of the APC class, correcting for which would increase the %Over for the most common APC champs.

Janna and Soraka aren't any more overused than Graves and Corki or Naut and Noc when one considers the size of their respective classes.


Comment below rating threshold, click here to show it.

fortide

Senior Member

06-11-2012

Mf and Caitlyn weren't picked at all...Guess their lane dominance niche wasn't enough


Comment below rating threshold, click here to show it.

FrankyCentaur

Senior Member

06-11-2012

I wish Riot would take a look at this and realize that 30% of champions in this game are useless ATM, and some of them only need some minor fixes -_-


Comment below rating threshold, click here to show it.

PallasAthens

Senior Member

06-11-2012

No Mordekaiser?

Sheesh, that's kinda sad he wasn't picked. A good team with Morde will wreck a team easy :l


Comment below rating threshold, click here to show it.

Krebstar

Senior Member

06-11-2012

Quote:
FrankyCentaur:
I wish Riot would take a look at this and realize that 30% of champions in this game are useless ATM, and some of them only need some minor fixes -_-


Yea, I feel exactly the same way... I think a lot more could be done for bottom lane in particular by buffing supports than by nerfing them. If the aggressive supports received more buffs, we might see more lane-domination champs like Cait and MF more often. I understand that balance is probably more complicated than I realize, but there seems to be some very obvious issues that never get dealt with while they do small things that have small effects (like tweaking Rammus' base stats, or lowering the ratio of Galio's Q... did anyone actually have a problem with either of these champs?)

Quote:
PallasAthens:
No Mordekaiser?

Sheesh, that's kinda sad he wasn't picked. A good team with Morde will wreck a team easy :l


Honestly, I might have missed him because of the way elobuff organizes AP champs (not by lane, but by whether they're a "bruiser&quot, but I don't recall seeing any mordes. This is the same reason I'm less confident in my APC results, but I think it's a good 1st look.


Comment below rating threshold, click here to show it.

xSoul

Senior Member

06-11-2012

Good research. Always pickin them a good suppot champ :P


Comment below rating threshold, click here to show it.

Deathwingduck

Senior Member

06-11-2012

Don't mind me. Just markin this page for when I'm on my computer later. Deathwingduck


Comment below rating threshold, click here to show it.

fel

Senior Member

06-11-2012

Anivia with an average of 0.4 death per games. GG froggen.


Comment below rating threshold, click here to show it.

jakab

Senior Member

06-11-2012

Malphite OP.


Comment below rating threshold, click here to show it.

Krebstar

Senior Member

06-11-2012

bump


12