Loading...
  OR  Zero-K Name:    Password:   
Title: [A] Teams All Welcome
Host: Nobody
Game version: Zero-K v1.13.9.5
Engine version: 2025.04.11
Battle ID: 2268275
Started: 11 days ago
Duration: 21 minutes
Players: 15
Bots: False
Mission: False
Rating: Casual
Watch Replay Now
Manual download


Show winners



Preview
Filter:    Player:  
sort

11 days ago
/random would do a better job at balancing teams. All low players on the same side is dumb as fuck.
+0 / -0
11 days ago
+1 / -0
What adjustment would you make to improve the team balancing algo?
+0 / -0
10 days ago
hard to know if a purple and a red are worth a blue and a silver if the elo is evenly balanced
+0 / -0
What is the current balancing algorithm? I can say for sure that if it treats elo as linear scale, then it is very, very wrong.

Edit: To be more explicit, what I'm saying is that if 1000+3000 elo is considered balanced against 2000+2000, then it is wrong.

Edit2: Treating 2050+1950 as balanced against 2000+2000 is good enough. It just breaks down at larger differences. I consider balancing for very large elo difference to be very hard. Ideally it could be avoided.
+0 / -0
There are probably multiple problems with the balancing algorithm, for example http://zero-k.info/Forum/Thread/36555 , to quote "The results indicate that the larger team does have a moderate advantage depending on team sizes and the exact model"

That's on top of any potential issue of rank distribution, because I think in this case the problem was that the distribution of the two teams was so "different" (basically "best ranked + lowest ranked" versus "middle ranked")

Now, I don't think that changing it without a through analysis is smart either. I have seen games that I thought they were worse balanced than this one and the result was opposite than what I thought initially...
+1 / -0