How does scoring work?
-
Trueskill was developed by Microsoft so I suggest we all contact Microsoft support and tell them to change the algorithm.
-
@ginandtonicbot said in How does scoring work?:
Thanks for the info. I get that. I'm now interested in understanding how mu is adjusted. i.e. same for all players on winning team, or different?
It is different.
Is there a simpler answer than a 57-page word doc?
Yes.
Seems like you're saying that even though no information is collected to assess player performance, players are given different mu adjustments.
The after game rating changes that you see are changes in displayed rating, which can basically be thought of as "base rating (mu) - uncertainty (3 * sigma) = displayed rating". The higher the sigma value, the lower the displayed rating and the more uncertain a rating is. So, new players start at "1500 base rating - 1500 uncertainty = 0 displayed rating". A typical player with a lot of games has around 250-300 rating uncertainty. Playing rated games lowers uncertainty to around 100-350 uncertainty as a stable level whose exact value largely depends on whether the player plays a lot of 1v1's (less uncertainty), 3v3's, 8v8's (more uncertainty), etc. Since players start with 1500 uncertainty, their ratings are initially more volatile and they more readily gain/lose rating. As they play more rated games, the uncertainty decreases and their ratings stabilize more, resulting in smaller rating changes from individual games. Additionally, if a player wins a game that was balanced in the player's favor, then the player gains a smaller amount of base rating (mu) than if the game was balanced against the player's favor (based on rating). Also, smaller game sizes (ie: 1v1's) generally have more impact than larger game sizes (ie: 8v8's).
