Help for package sport

Type:

Package

Title:

Sequential Pairwise Online Rating Techniques

Version:

0.2.1

Depends:

R (≥ 3.0)

Maintainer:

Dawid Kałędkowski <dawid.kaledkowski@gmail.com>

Description:

Calculates ratings for two-player or multi-player challenges. Methods included in package such as are able to estimate ratings (players strengths) and their evolution in time, also able to predict output of challenge. Algorithms are based on Bayesian Approximation Method, and they don't involve any matrix inversions nor likelihood estimation. Parameters are updated sequentially, and computation doesn't require any additional RAM to make estimation feasible. Additionally, base of the package is written in C++ what makes sport computation even faster. Methods used in the package refer to Mark E. Glickman (1999) http://www.glicko.net/research/glicko.pdf; Mark E. Glickman (2001) <doi:10.1080/02664760120059219>; Ruby C. Weng, Chih-Jen Lin (2011) https://www.jmlr.org/papers/volume12/weng11a/weng11a.pdf; W. Penny, Stephen J. Roberts (1999) <doi:10.1109/IJCNN.1999.832603>.

BugReports:

https://github.com/gogonzo/sport/issues

Imports:

Rcpp, data.table, ggplot2

LinkingTo:

Rcpp

License:

GPL-2

Encoding:

UTF-8

Language:

en-US

URL:

https://github.com/gogonzo/sport

LazyData:

true

RoxygenNote:

7.2.3

Suggests:

dplyr, knitr, lobstr, rmarkdown, testthat

VignetteBuilder:

knitr

NeedsCompilation:

yes

Packaged:

2024-01-08 16:25:24 UTC; gonzo

Author:

Dawid Kałędkowski

[aut, cre]

Repository:

CRAN

Date/Publication:

2024-01-08 23:50:02 UTC

Bayesian Bradley-Terry

Description

Bayesian Bradley-Terry

Usage

bbt_run(
  formula,
  data,
  r = numeric(0),
  rd = numeric(0),
  init_r = 25,
  init_rd = 25/3,
  lambda = NULL,
  share = NULL,
  weight = NULL,
  kappa = 0.5
)

Arguments

formula

formula which specifies the model. RHS Allows only player rating parameter and it should be specified in following manner:

rank | id ~ player(name).

rank player position in event.
id event identifier in which pairwise comparison is assessed.
player(name) name of the contestant. In this case player(name) helps algorithm point name of the column where player names are stored.

Users can also specify formula in in different way:

rank | id ~ player(name|team). Which means that players are playing in teams, and results are observed for teams not for players. For more see vignette.

data

data.frame which contains columns specified in formula, and optional columns defined by lambda, weight.

r

named vector of initial players ratings estimates. If not specified then r will be created automatically for parameters specified in formula with initial value init_r.

rd

rd named vector of initial rating deviation estimates. If not specified then rd will be created automatically for parameters specified in formula with initial value init_rd.

init_r

initial values for r if not provided. Default (glicko = 1500, glicko2 = 1500, bbt = 25, dbl = 0)

init_rd

initial values for rd if not provided. Default (glicko = 350, glicko2 = 350, bbt = 25/3, dbl = 1)

lambda

name of the column in data containing lambda values or one constant value (eg. lambda = colname or lambda = 0.5). Lambda impact prior variance, and uncertainty of the matchup result. The higher lambda, the higher prior variance and more uncertain result of the matchup. Higher lambda flattens chances of winning.

share

name of the column in data containing player share in team efforts. It's used to first calculate combined rating of the team and then redistribute ratings update back to players level. Warning - it should be used only if formula is specified with players nested within teams (player(player|team)).

weight

name of the column in data containing weights values or one constant (eg. weight = colname or weight = 0.5). Weights increasing (weight > 1) or decreasing (weight < 1) update change. Higher weight increasing impact of event result on rating estimate.

kappa

controls rd shrinkage not to be greater than rd*(1 - kappa). kappa=1 means that rd will not be decreased.

Value

A "rating" object is returned:

final_r named vector containing players ratings.
final_rd named vector containing players ratings deviations.
r data.frame with evolution of the ratings and ratings deviations estimated at each event.
pairs pairwise combinations of players in analysed events with prior probability and result of a challenge.
class of the object.
method type of algorithm used.
settings arguments specified in function call.

Examples

# the simplest example
data <- data.frame(
  id = c(1, 1, 1, 1),
  team = c("A", "A", "B", "B"),
  player = c("a", "b", "c", "d"),
  rank_team = c(1, 1, 2, 2),
  rank_player = c(3, 4, 1, 2)
)

bbt <- bbt_run(
  data = data,
  formula = rank_player | id ~ player(player),
  r = setNames(c(25, 23.3, 25.83, 28.33), c("a", "b", "c", "d")),
  rd = setNames(c(4.76, 0.71, 2.38, 7.14), c("a", "b", "c", "d"))
)

# nested matchup
bbt <- bbt_run(
  data = data,
  formula = rank_team | id ~ player(player | team)
)

Dynamic Bayesian Logit

Description

Dynamic Bayesian Logit

Usage

dbl_run(
  formula,
  data,
  r = NULL,
  rd = NULL,
  lambda = NULL,
  weight = NULL,
  kappa = 0.95,
  init_r = 0,
  init_rd = 1
)

Arguments

formula

formula which specifies the model. Unlike other algorithms in the packages (glicko_run, glicko2_run, bbt_run), this method doesn't allow players nested in teams with player(player | team) and user should matchup in formula using player(player). DBL allows user specify multiple parameters also in interaction with others.

data

data.frame which contains columns specified in formula, and optional columns defined by lambda, weight.

r

named vector of initial players ratings estimates. If not specified then r will be created automatically for parameters specified in formula with initial value init_r.

rd

rd named vector of initial rating deviation estimates. If not specified then rd will be created automatically for parameters specified in formula with initial value init_rd.

lambda

weight

kappa

controls rd shrinkage not to be greater than rd*(1 - kappa). kappa=1 means that rd will not be decreased.

init_r

initial values for r if not provided. Default (glicko = 1500, glicko2 = 1500, bbt = 25, dbl = 0)

init_rd

initial values for rd if not provided. Default (glicko = 350, glicko2 = 350, bbt = 25/3, dbl = 1)

Value

A "rating" object is returned:

final_r named vector containing players ratings.
final_rd named vector containing players ratings deviations.
r data.frame with evolution of the ratings and ratings deviations estimated at each event.
pairs pairwise combinations of players in analysed events with prior probability and result of a challenge.
class of the object.
method type of algorithm used.
settings arguments specified in function call.

Examples

# the simplest example

data <- data.frame(
  id = c(1, 1, 1, 1),
  name = c("A", "B", "C", "D"),
  rank = c(3, 4, 1, 2),
  gate = c(1, 2, 3, 4),
  factor1 = c("a", "a", "b", "b"),
  factor2 = c("a", "b", "a", "b")
)

dbl <- dbl_run(
  data = data,
  formula = rank | id ~ player(name)
)

dbl <- dbl_run(
  data = data,
  formula = rank | id ~ player(name) + gate * factor1
)

Glicko2 rating algorithm

Description

Glicko2 rating algorithm

Usage

glicko2_run(
  formula,
  data,
  r = numeric(0),
  rd = numeric(0),
  sigma = numeric(0),
  lambda = NULL,
  share = NULL,
  weight = NULL,
  init_r = 1500,
  init_rd = 350,
  init_sigma = 0.05,
  kappa = 0.5,
  tau = 0.5
)

Arguments

formula

formula which specifies the model. RHS Allows only player rating parameter and it should be specified in following manner:

rank | id ~ player(name).

rank player position in event.
id event identifier in which pairwise comparison is assessed.
player(name) name of the contestant. In this case player(name) helps algorithm point name of the column where player names are stored.

Users can also specify formula in in different way:

rank | id ~ player(name|team). Which means that players are playing in teams, and results are observed for teams not for players. For more see vignette.

data

data.frame which contains columns specified in formula, and optional columns defined by lambda, weight.

r

named vector of initial players ratings estimates. If not specified then r will be created automatically for parameters specified in formula with initial value init_r.

rd

rd named vector of initial rating deviation estimates. If not specified then rd will be created automatically for parameters specified in formula with initial value init_rd.

sigma

(only for glicko2) named vector of initial players ratings estimates. If not specified then sigma will be created automatically for parameters specified in formula with initial value init_sigma.

lambda

share

weight

init_r

initial values for r if not provided. Default (glicko = 1500, glicko2 = 1500, bbt = 25, dbl = 0)

init_rd

initial values for rd if not provided. Default (glicko = 350, glicko2 = 350, bbt = 25/3, dbl = 1)

init_sigma

initial values for sigma if not provided. Default = 0.5

kappa

controls rd shrinkage not to be greater than rd*(1 - kappa). kappa=1 means that rd will not be decreased.

tau

The system constant. Which constrains the change in volatility over time. Reasonable choices are between 0.3 and 1.2 (default = 0.5), though the system should be tested to decide which value results in greatest predictive accuracy. Smaller values of tau prevent the volatility measures from changing by large amounts, which in turn prevent enormous changes in ratings based on very improbable results. If the application of Glicko-2 is expected to involve extremely improbable collections of game outcomes, then tau should be set to a small value, even as small as, say, tau= 0.

Value

A "rating" object is returned:

final_r named vector containing players ratings.
final_rd named vector containing players ratings deviations.
final_sigma named vector containing players ratings volatile.
r data.frame with evolution of the ratings and ratings deviations estimated at each event.
pairs pairwise combinations of players in analysed events with prior probability and result of a challenge.
class of the object.
method type of algorithm used.
settings arguments specified in function call.

Examples

# the simplest example
data <- data.frame(
  id = c(1, 1, 1, 1),
  team = c("A", "A", "B", "B"),
  player = c("a", "b", "c", "d"),
  rank_team = c(1, 1, 2, 2),
  rank_player = c(3, 4, 1, 2)
)

# Example from Glickman
glicko2 <- glicko2_run(
  data = data,
  formula = rank_player | id ~ player(player),
  r = setNames(c(1500.0, 1400.0, 1550.0, 1700.0), c("a", "b", "c", "d")),
  rd = setNames(c(200.0, 30.0, 100.0, 300.0), c("a", "b", "c", "d"))
)

# nested matchup
glicko2 <- glicko2_run(
  data = data,
  formula = rank_team | id ~ player(player | team)
)

Glicko rating algorithm

Description

Glicko rating algorithm

Usage

glicko_run(
  data,
  formula,
  r = numeric(0),
  rd = numeric(0),
  init_r = 1500,
  init_rd = 350,
  lambda = numeric(0),
  share = numeric(0),
  weight = numeric(0),
  kappa = 0.5
)

Arguments

data

data.frame which contains columns specified in formula, and optional columns defined by lambda, weight.

formula

formula which specifies the model. RHS Allows only player rating parameter and it should be specified in following manner:

rank | id ~ player(name).

rank player position in event.
id event identifier in which pairwise comparison is assessed.
player(name) name of the contestant. In this case player(name) helps algorithm point name of the column where player names are stored.

Users can also specify formula in in different way:

rank | id ~ player(name|team). Which means that players are playing in teams, and results are observed for teams not for players. For more see vignette.

r

named vector of initial players ratings estimates. If not specified then r will be created automatically for parameters specified in formula with initial value init_r.

rd

rd named vector of initial rating deviation estimates. If not specified then rd will be created automatically for parameters specified in formula with initial value init_rd.

init_r

initial values for r if not provided. Default (glicko = 1500, glicko2 = 1500, bbt = 25, dbl = 0)

init_rd

initial values for rd if not provided. Default (glicko = 350, glicko2 = 350, bbt = 25/3, dbl = 1)

lambda

share

weight

kappa

controls rd shrinkage not to be greater than rd*(1 - kappa). kappa=1 means that rd will not be decreased.

Value

A "rating" object is returned:

final_r named vector containing players ratings.
final_rd named vector containing players ratings deviations.
r data.frame with evolution of the ratings and ratings deviations estimated at each event.
pairs pairwise combinations of players in analysed events with prior probability and result of a challenge.
class of the object.
method type of algorithm used.
settings arguments specified in function call.

Examples

# the simplest example
data <- data.frame(
  id = c(1, 1, 1, 1),
  team = c("A", "A", "B", "B"),
  player = c("a", "b", "c", "d"),
  rank_team = c(1, 1, 2, 2),
  rank_player = c(3, 4, 1, 2)
)

# Example from Glickman
glicko <- glicko_run(
  data = data,
  formula = rank_player | id ~ player(player),
  r = setNames(c(1500.0, 1400.0, 1550.0, 1700.0), c("a", "b", "c", "d")),
  rd = setNames(c(200.0, 30.0, 100.0, 300.0), c("a", "b", "c", "d"))
)

# nested matchup
glicko <- glicko_run(
  data = data,
  formula = rank_team | id ~ player(player | team)
)

Heat results of Speedway Grand-Prix

Description

Actual dataset containing heats results of all Speedway Grand-Prix turnaments gpheats.

Format

A data frame with >19000 rows and 11 variables:

id - event identifier
season - year of Grand-Prix, 1995-now
date - date of turnament
round - round in season
name - Turnament name
heat - heat number, 1-23
field - number of gate, 1-4
rider - rider name, string
points - paints gained, integer
position - position at finish line, string
rank - rank at finish line, integer

Source

internal

Turnament results of Speedway Grand-Prix

Description

Actual dataset containing turnament results of all Speedway Grand-Prix events gpsquads

Format

A data frame with >4000 rows and 9 variables:

id - event identifier
season - year of Grand-Prix, 1995-now
date - date of turnament
place - stadium of event
round - round in season
name - Turnament name
rider - rider names, 1-6
points - points gained, integer
classification - classification after an event

Source

internal

Plot rating object

Description

Plot rating object

Usage

## S3 method for class 'rating'
plot(x, n = 10, players, ...)

Arguments

x

of class rating

n

number of teams to be plotted

players

optional vector with names of the contestants (coefficients) to plot their evolution in time.

...

optional arguments

Predict rating model

Description

Predict rating model

Usage

## S3 method for class 'rating'
predict(object, newdata, ...)

Arguments

object

of class rating

newdata

data.frame with data to predict

...

optional arguments

Value

probabilities of winning challenge by player over his opponent in all provided events.

Examples

glicko <- glicko_run(
  data = gpheats[1:16, ],
  formula = rank | id ~ player(rider)
)
predict(glicko, gpheats[17:20, ])

Apply rating algorithm

Description

Apply rating algorithm

Usage

rating_run(
  method,
  data,
  formula,
  r = numeric(0),
  rd = numeric(0),
  sigma = numeric(0),
  init_r = numeric(0),
  init_rd = numeric(0),
  init_sigma = numeric(0),
  lambda = numeric(0),
  share = numeric(0),
  weight = numeric(0),
  kappa = numeric(0),
  tau = numeric(0)
)

Arguments

method

one of c("glicko", "glicko2", "bbt", "dbl")

data

data.frame which contains columns specified in formula, and optional columns defined by lambda, weight.

formula

formula which specifies the model. RHS Allows only player rating parameter and it should be specified in following manner:

rank | id ~ player(name).

rank player position in event.
id event identifier in which pairwise comparison is assessed.
player(name) name of the contestant. In this case player(name) helps algorithm point name of the column where player names are stored.

Users can also specify formula in in different way:

rank | id ~ player(name|team). Which means that players are playing in teams, and results are observed for teams not for players. For more see vignette.

r

named vector of initial players ratings estimates. If not specified then r will be created automatically for parameters specified in formula with initial value init_r.

rd

rd named vector of initial rating deviation estimates. If not specified then rd will be created automatically for parameters specified in formula with initial value init_rd.

sigma

init_r

initial values for r if not provided. Default (glicko = 1500, glicko2 = 1500, bbt = 25, dbl = 0)

init_rd

initial values for rd if not provided. Default (glicko = 350, glicko2 = 350, bbt = 25/3, dbl = 1)

init_sigma

initial values for sigma if not provided. Default = 0.5

lambda

share

weight

kappa

controls rd shrinkage not to be greater than rd*(1 - kappa). kappa=1 means that rd will not be decreased.

tau

Summarizing rating objects

Description

Summarizing rating objects Summary for object of class rating

Usage

## S3 method for class 'rating'
summary(object, ...)

Arguments

object

of class rating

...

optional arguments

Value

List with following elements:

formula modeled formula.
method type of algorithm used.
⁠Overall Accuracy⁠ named vector containing players ratings.
r a data.frame with summarized players ratings and model winning probabilities.

Probabilities are returned only in models with one variable (ratings):

name of a player
r players ratings
rd players ratings deviation
⁠Model probability⁠ mean predicted probability of winning the challenge by the player.
⁠True probability⁠ mean observed probability of winning the challenge by the player.
Accuracy Accuracy of prediction.
pairings number of pairwise occurrences.

Examples

model <- glicko_run(
  formula = rank | id ~ player(rider),
  data = gpheats[1:102, ]
)
summary(model)