PlayerRatings: plot.rating – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

plot.rating

Plot Player Features Across Time for a Rating Object

Description

Plot line traces of estimated ratings or other features for selected players. This function can only be used if the full history is retained in the object x.

Usage

## S3 method for class 'rating'
plot(x, which = "Rating", players = NULL, t0 = 1, tv = NULL,
  npl = 10, random = FALSE, xlab = "Time Period", ylab = paste(x$type," Ratings"), 
  main = paste(x$type," Ratings System"), inflation = FALSE, add=FALSE, ...)

Arguments

`x`	An object of class `"rating"`.
`which`	The variable to be plotted.
`players`	If not `NULL`, should be a vector of player identifiers to explicitly select players to be plotted.
`t0`	The time index at which to begin. Note that unless players are specified explicitly, players who do not play at time index `t0` will not be selected for the plot. Can also be a vector of length two, in which case the second value is the time index at which to end.
`tv`	If not `NULL`, then a vector of values to be used on the x-axis instead of the time index.
`npl`	The number of players to select.
`random`	If `TRUE`, `npl` players are selected at random from those who played at time `t0`. If `FALSE` (the default), the `npl` players who played most games at `t0` are selected. Ignored if `players` is not `NULL`.
`xlab,ylab,main`	Graphical parameters.
`inflation`	If `TRUE`, plot the average rating of the best `npl` players at each time point. This is designed to investigate ratings inflation.
`add`	Add to an existing plot.
`...`	Other parameters to be passed through to plotting functions.

Details

Note that the argument random is not used by default, since it can produce flat profiles from randomly selected players who play few games. The default selection is non-random and selects more active players, however they may be more likely to improve over time than the general population.

Examples

afl <- aflodds[,c(2,3,4,7)]
robj <- steph(afl, history=TRUE)
plot(robj)

robj <- elom(riichi, history = TRUE)
pl <- robj$ratings$Player[robj$ratings$Games >= 80]
plot(robj, players = pl)

PlayerRatings

Dynamic Updating Methods for Player Ratings Estimation

v1.1-0

GPL-3

Authors

Alec Stephenson and Jeff Sonas.

Initial release

2020-02-28