Brawlhalla

Brawlhalla

View Stats:
DesuTronic Jul 18, 2018 @ 12:31pm
I harvested all the data about global player elo distribution. Enjoy!
Last edited by DesuTronic; Jul 18, 2018 @ 12:32pm
< >
Showing 1-5 of 5 comments
Lychee Jul 18, 2018 @ 12:34pm 
Neat!
DesuTronic Jul 18, 2018 @ 2:14pm 
Originally posted by kuvasz:
That can't be right lol
I think the f2p model skews it a lot. Thanks for your efforts though :VolgarrChalice:
f2p model... what?
Been awhile since I've seen a high quality post here...congrats
DesuTronic Jul 18, 2018 @ 3:40pm 
Originally posted by kuvasz:
Originally posted by DesuTronic:
f2p model... what?
The free to play model greatly contributes to elo inflation via abandoned accounts and the skewing of elo distribution via shorter average playtime/skill per account.
That is specifically why I stated in the comments with a long post about how I removed all accounts that have played less than 10 ranked games and even made a whole seperate chart only including accounts with 50 games or more.
And much more, please go read the whole thing.
Last edited by DesuTronic; Jul 18, 2018 @ 3:41pm
DesuTronic Jul 18, 2018 @ 4:17pm 
Originally posted by kuvasz:
Originally posted by DesuTronic:
That is specifically why I stated in the comments with a long post about how I removed all accounts that have played less than 10 ranked games and even made a whole seperate chart only including accounts with 50 games or more.
And much more, please go read the whole thing.
Woah triggered much?
I was just stating a general truth about the business model affecting the numbers. Admittedly I only checked the first image on my phone without reading the text but when you have such low threasholds it doesn't matter. From what I can tell, the 50-game category is supposed to be your stricter but more reliable group... when 50 games is a mere 4 hours of gameplay time. Let's assume the player doesn't jump right into ranked but instead practices hard... let's be generous and give him 10 hours of game time. That's 2.5 times your allocation and it's still extremely low, I probably even got lost in the menu after just 10 hours.
And the output for "who have played quite a bit more games" is just not inviting or easy to interpret, sorry.

I did jump the gun but the presentation needs work.
The ranking list resets with every season so that graph only has the last month worth of players in it.
I chose 10 games played because thats the point where someone qualifies for the glory reward so most would stop at that point. Just by removing the players with 10 or less games played already removes 60% of the players from the list (from 280k to 113k)
And then you say that even 50 played games is "mere 4 hours", that 50 wins or more chart only has 17% of the remaining playerbase left.
What would be an acceptable cuttoff point for you? Anything over 100 games would be a "mere 8 hours" but would practically be only 5% of the actual playerbase.
If I'd used all accounts that have ever been created in the several years of brawlhalla then yes I'd agree that something like 100 or 200 games would be a minimum or at least deserving to be the main chart but there's only a few thousand players at best who have played over 100 games in the past month.
< >
Showing 1-5 of 5 comments
Per page: 1530 50

Date Posted: Jul 18, 2018 @ 12:31pm
Posts: 5