Loading...
  OR  Zero-K Name:    Password:   

Post edit history

Evaluating rating systems

To display differences between versions, select one or more edits in the list using checkboxes and click "diff selected"
Post edit history
Date Editor Before After
9/2/2016 3:34:52 AMCHrankAdminDeinFreund before revert after revert
Before After
1 I'm using all publicly available battles for the testing.
2 \n
1 Using sqrt(N/2) factor for teamstrength: 3 Using sqrt(N/2) factor for teamstrength:
2 {{{ 4 {{{
3 Teamstrength with Coms: 0.0214564709657561 5 Teamstrength with Coms: 0.0214564709657561
4 Teamstrength without Coms: 0.02153829204252216 6 Teamstrength without Coms: 0.02153829204252216
5 Teamstrength with Coms (K=64): 0.013549012710202778 7 Teamstrength with Coms (K=64): 0.013549012710202778
6 Teamstrength without Coms (K=64): 0.013519467431481061 8 Teamstrength without Coms (K=64): 0.013519467431481061
7 ZK Elo: 0.023980837112044084 9 ZK Elo: 0.023980837112044084
8 ZK Elo (K=64): 0.02683235735477585 10 ZK Elo (K=64): 0.02683235735477585
9 }}} 11 }}}
12 \n
13 I've been playing around with D as well, but it seems to do pretty much the same as the K factor mod. If I combine both D and K mod the results go negative.. Wasn't there some idea of a fundamental improvement over ELO? I wonder if we could get to try some actually different rating systems.