elo v3.0.2

elo v3.0.1

Breaking changes:

Restricted the version of R to >= 3.6.
Changed the impact of the group() formula special in elo.run(). It now indicates when to update Elos. (#54)
Removed elo.run2(), to be replaced by passing arguments to elo.run(). A message is issued when the R backend is used.
The [.elo.k() (new) and [.elo.players.matrix() methods now drop any extra classes when j= is specified.
Changed a warning to an error in final.elos() when using regressed=TRUE without regression after the last game.
elo.glm(), elo.markovchain(), elo.winpct(), elo.colley() now emit NAs for running predictions on groups that haven’t been seen yet. (#56)

Other changes:

Added subset= argument to auc() and favored().
Added ignore.skipped=FALSE argument to auc(), favored(), and mse() running methods.
Added attributes to fitted.elo.running(..., running=TRUE) to indicate the group (i.e., model) from which the prediction arises.
Added elo.run.multiteam() for when matchups consist of multiple teams. (#54)
Improved documentation, including expanding from 1 to 3 vignettes
Fixed a bug with the as.data.frame() method for elo.run() when players() are involved. (#55)
Changed how tournament is created (though the data didn’t actually change).
Made some methods more explicit: length.elo.k(), is.na.elo.k(), [.elo.k(), is.na.elo.players.matrix()
Made one fix for R-devel related to subsetting a vector with a classed object.

Added elo.colley(), with its corresponding helper functions.
Allowed k() to take two arguments, to give differential updates to “team.A” and “team.B”. This has one user-visible effect: as.data.frame.elo.run() now has one more column than it did before, and its column names have changed. (#45)
Added elo.run2(), which allows for custom probabilities and updates, but by default returns the same as elo.run() (except more slowly). (#47)
Added a pkgdown site: https://eheinzen.github.io/elo/

Added the running=TRUE option to elo.glm(). This gives an object of class "elo.running", with corresponding methods for summary(), fitted(), predict(), mse(), auc(), and favored().
Added weights= to elo.glm().
Added support for adjust() in elo.glm() to include adjustments in the logistic regression.
Added a new inline function neutral(), to denote neutral field in elo.glm() and elo.markovchain().
Removed the rm.ties= argument from elo.glm(). Ties will have to be removed instead with subset= or before running the function altogether.
Added elo.markovchain(), with corresponding methods for summary(), fitted(), predict(), mse(), auc(), and favored(). This also has the running=TRUE option.
Added elo.winpct(), with corresponding methods for summary(), fitted(), predict(), mse(), auc(), and favored(). This also has the running=TRUE option.
Added a function to denote margin of victory, for continuous modeling in elo.glm(), elo.markovchain(), and elo.winpct(): mov().

Fixed a bug with adding NAs back in to fitted values and residuals with na.exclude() in elo.glm() and elo.run(). (#39, #42)
Fixed a bug with adjust() variables not getting subsetted correctly with na.action in model.frame(). (#40)
Added is.na.elo.adjust() to test for NAs in the adjustment vector. (#41)

Widened the version dependency to R 3.3.0.
Allowed players() matrices in elo.run() to find Elos of individual players playing at the same time.
Added elo.glm(), a simple function to run logistic regressions on Elo setups.
Fixed a bug in the favored() function (used in summary.elo.run()). (#29)
Exported and revamped the class structure of the specials allowed in formulas. (#30)
Allowed access to elo.model.frame() even when the package isn’t loaded. (#34)
Allowed regression to different values for each team. (#35)

Fixed a bug with initial Elos and deep copying in C++. (#25)
Added an argument to regress() allowing users to stop regressing teams which have stopped playing. (#26)

This version is not backwards compatible!

Changed the signatures of elo.calc() and elo.update() to match formula interface.
Changed elo.calc(), elo.update(), and elo.prob() to S3 generics, and implemented formula methods. The default methods now include options to adjust Elos. (#3)
elo.run():
- elo.run() no longer accepts numeric values for team.A.
- elo.run() now accepts special functions group() and regress(). If the latter is used, the class of the returned object becomes "elo.run.regressed". (#11, #12, #19, #22)
- The $elos component of "elo.run" objects has been completely reworked, and now uses 1-based indexing. Because of this, the print.elo.run() method also had to be fixed. (#16)
Renamed last() to final.elos() (#9).
Changed tournament dataset.

The elo package now imports pROC::auc().
elo.prob() now accepts vectors of team names (like elo.run()) as input. (#6)
Documentation and the vignette have been updated.

Implemented elo.model.frame(). The output is a data.frame with appropriately named columns.
Implemented predict.elo.run() and predict.elo.run.regressed(). (#2, #19)
Added is.score() to test for “score-ness”.
Implemented summary.elo.run(), along with helpers to calculate AUC and MSE (auc() and mse()). (#15)