Re:Elo sucks – better multiplayer rating systems for smaller...
by IshKebab on Tuesday July 21, 2020 @05:27PM
(#23911195)
TrueSkill definitely has a time decay term and I'm fairly sure it lets you fit the model to previous games. I wonder if the author actually tried it. (Though to be fair I'm not sure if there are open source versions of the latest version of TrueSkill.)
[ Parent
| on HN ]
Re:Elo sucks – better multiplayer rating systems for smaller...
by BSTRhino on Tuesday July 21, 2020 @07:25PM
(#23912026)
Yes, tried Glicko then TrueSkill, both generated huge amounts of complaints. New system produced few complaints. If the community had liked it, would've stuck with TrueSkill.
Re:Elo sucks – better multiplayer rating systems for smaller...
by IshKebab on Wednesday July 22, 2020 @10:44AM
(#23917556)
TrueSkill 1 presumably?
Re:Elo sucks – better multiplayer rating systems for smaller...
by karlding on Tuesday July 21, 2020 @08:02PM
(#23912327)
The parent is talking about TrueSkill 2 [0], while the trueskill Python library you linked currently only supports the original TrueSkill algorithm [1][2]. TrueSkill 2 takes into account individual scores of players in order to weigh the contribution of each player to each team. The idea is that this allows TrueSkill 2 to converge faster for new players.
[0] https://www.microsoft.com/en-us/research/publication/trueski...
[1] https://github.com/sublee/trueskill/issues/27
[2] https://www.microsoft.com/en-us/research/project/trueskill-r...