Vixen
caa30e12
csharp
public interface IRatingModel

How ratings change and how well two sides are matched.

Read the guide page for this →

Remarks

Doc 28 ships two and picks neither. Elo is one number, transparent, and right for 1v1 and for games whose players will ask how it works. A Bayesian model of the TrueSkill family carries a variance, which is what handles teams, parties, uneven sizes and new players honestly. The queue definition chooses.

Fields and properties (2)

  • string Name

    What it is called, in a queue definition and in a report.

  • Rating Starting

    What somebody starts on.

Methods (2)

  • IReadOnlyList<Rating[]> Update(IReadOnlyList<Rating[]> teams, IReadOnlyList<int> ranks)

    Updates every team's ratings after a result.

  • double Quality(IReadOnlyList<Rating[]> teams)

    How even a match would be, from zero to one.

Used by (3)

  • BayesianRatingModelVixen.Live.Matchmaking
  • DuelsVixen.Live.Matchmaking.Tests
  • EloRatingModelVixen.Live.Matchmaking