Register    Login    Forum    Search    FAQ Awesomenauts



Post new topic Reply to topic

Author Message
 Post subject: Should bot win percentage influence rating?
 Post Posted: Tue Jun 30, 2015 12:45 am 
Offline
User avatar

Joined: Thu Aug 08, 2013 12:59 am
Posts: 1598
Saw this one reddit

Image

Ignoring the obvious solar bug, this got me thinking about how badly the rating system works for these kinds of situations. While the best cure would be something significantly more nuanced than simple win/lose, perhaps it might help to give bots different ratings depending on how they perform. For example if, say, Clunkbot has a win percentage that is 60% higher than Skolbot's, wouldn't it make sense to assign him a higher rating? To do so would increase the accuracy of the rating system when bots are in a match.

It's also a bit odd that bots gain additional cheats in higher leagues while their rating stays the same. A bot with BotLevel 2 cheats is significantly better than one with BetLevel 0, so why are they rated equally? Ideally bots should be ranked higher in upper leagues and lower in lesser leagues. If the rating remains static, then you're misrepresenting the challenge bots pose on at least one of these levels.


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Tue Jun 30, 2015 1:29 am 
Offline
User avatar

Joined: Sat Sep 07, 2013 5:56 pm
Posts: 539
Location: Somewhere in a Canadian laboratory
I've always liked the idea of giving bots a "spot" on the leaderboards, for them to move around as they win/lose games, and then the bots get assigned to matches based on league. Would also provide niches to fill for people looking to join the community, such as "create a simple froggy G bot for league 7". That's an ideal situation for me, but I can get behind your idea as well, Hurley.

_________________
Myah!

Bots I've made:

[In-Game] Derpl viewtopic.php?f=23&t=35642 :drool:

http://imgur.com/5P0Ais3


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Tue Jun 30, 2015 3:10 am 
Offline
User avatar

Joined: Thu Aug 08, 2013 12:59 am
Posts: 1598
Your idea sounds awesome. You could then try to even out teams with bots on both sides by assigning different bots based on the average rating of the players on that team. Sounds like a lot of work though.


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Tue Jun 30, 2015 2:27 pm 
Offline
User avatar

Joined: Wed Jul 30, 2014 3:15 pm
Posts: 1160
Location: 3 miles under in a text file
I agree with Mesp. That would also make it easy to tell which bots need improvement. Those in the lower leagues would be the worst performers. Of course, that system isn't perfect. Some 'Nauts (Yuri comes to mind) are weaker than most. Ronimo should also buff such 'Nauts, ideally before considering removing the bot.

_________________
"If it works well the first time, something is really, really wrong."

Veanko wrote:
Ignotis the hedgehog confirmed


STEAM


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Fri Jul 03, 2015 9:39 am 
Offline
User avatar

Joined: Mon Sep 03, 2012 4:05 pm
Posts: 3315
Location: Melbourne, Australia
TheMesp wrote:
I've always liked the idea of giving bots a "spot" on the leaderboards, for them to move around as they win/lose games, and then the bots get assigned to matches based on league. Would also provide niches to fill for people looking to join the community, such as "create a simple froggy G bot for league 7". That's an ideal situation for me, but I can get behind your idea as well, Hurley.

didn't joost say they didn't want bot modding to be a competition or something...?
like, he didn't want to reveal win percentages of bots so people didn't feel bad or something. even though bots that had a low win % were kicked out of the main game, lol.

_________________
TigerKirby215 wrote:
People in L3 are a very tough challenge. So one can easily assume L1 live in insane asylums & have nothing to do other than play nauts, or hack
Quote:
Frankie3nig has been permabanned by Sony 0


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Fri Jul 03, 2015 1:34 pm 
Offline
User avatar

Joined: Wed Jul 30, 2014 3:15 pm
Posts: 1160
Location: 3 miles under in a text file
Dr. DaMan wrote:
TheMesp wrote:
I've always liked the idea of giving bots a "spot" on the leaderboards, for them to move around as they win/lose games, and then the bots get assigned to matches based on league. Would also provide niches to fill for people looking to join the community, such as "create a simple froggy G bot for league 7". That's an ideal situation for me, but I can get behind your idea as well, Hurley.

didn't joost say they didn't want bot modding to be a competition or something...?
like, he didn't want to reveal win percentages of bots so people didn't feel bad or something. even though bots that had a low win % were kicked out of the main game, lol.


We don't need to know the win percentages for this to work. Besides, the modding community is tightly-knit enough that competition wouldn't be a problem.

_________________
"If it works well the first time, something is really, really wrong."

Veanko wrote:
Ignotis the hedgehog confirmed


STEAM


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Fri Jul 03, 2015 1:36 pm 
Offline
User avatar

Joined: Mon Sep 02, 2013 5:16 pm
Posts: 1605
Location: Dænmårk
Ignotis wrote:
Besides, the modding community is tightly-knit enough that competition wouldn't be a problem.


:thumb:
I wouldn't mind competition at all. We're friendly people!


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Fri Jul 03, 2015 8:09 pm 
Offline
User avatar

Joined: Wed Jul 30, 2014 3:15 pm
Posts: 1160
Location: 3 miles under in a text file
FlawlessHair wrote:
Ignotis wrote:
Besides, the modding community is tightly-knit enough that competition wouldn't be a problem.


:thumb:
I wouldn't mind competition at all. We're friendly people!


It happened before with Foxless vs. Unstoppable Fox, and no feelings were harmed. We actually ended up helping each other. In other cases, Veanko didn't submit his Sentry bot to make way for mine, and I un-submitted my Genji bot to help trinary get a golden nutter. (no, Mesp, it's not a gearlyfish)

_________________
"If it works well the first time, something is really, really wrong."

Veanko wrote:
Ignotis the hedgehog confirmed


STEAM


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Sat Jul 04, 2015 2:42 pm 
Offline
User avatar

Joined: Sat Sep 07, 2013 5:56 pm
Posts: 539
Location: Somewhere in a Canadian laboratory
Ignotis wrote:
(no, Mesp, it's not a gearlyfish)


IT IS A GEARLYFISH.

I PUT IN BLOOD, SWEAT AND TEARS TO GET IT CALLED THAT.

Image

_________________
Myah!

Bots I've made:

[In-Game] Derpl viewtopic.php?f=23&t=35642 :drool:

http://imgur.com/5P0Ais3


Top 
 Profile  
 
 Post subject: Re: Should bot win percentage influence rating?
 Post Posted: Sat Jul 04, 2015 3:57 pm 
Offline
User avatar

Joined: Mon Sep 03, 2012 4:05 pm
Posts: 3315
Location: Melbourne, Australia
shadup you nutter

_________________
TigerKirby215 wrote:
People in L3 are a very tough challenge. So one can easily assume L1 live in insane asylums & have nothing to do other than play nauts, or hack
Quote:
Frankie3nig has been permabanned by Sony 0


Top 
 Profile  
 
Display posts from previous:  Sort by  
 
Post new topic Reply to topic

cron