`R/elo.glm.R`

`elo.glm.Rd`

Compute a (usually logistic) regression model for a series of matches.

elo.glm( formula, data, family = "binomial", weights, na.action, subset, ..., running = FALSE, skip = 0 )

formula | A formula. See the help page for formulas for details. |
---|---|

data | A |

family, weights, ... | Arguments passed to |

na.action | A function which indicates what should happen when the data contain NAs. |

subset | An optional vector specifying a subset of observations. |

running | Logical, denoting whether to calculate "running" projected probabilities. If true, a model is fit for
group 1 on its own to predict group 2, then groups 1 and 2 to predict 3, then groups 1 through 3 to predict 4, etc.
Groups are determined in |

skip | Integer, denoting how many groups to skip before fitting the running models. This is helpful if groups are small, where glm would have trouble converging for the first few groups. The predicted values are then set to 0.5 for the skipped groups. |

An object of class `c("elo.glm", "glm")`

. If `running==TRUE`

, the class `"elo.glm.running"`

is prepended.

The formula syntax is the same as other `elo`

functions. A data.frame
of indicator variables is built, where an entry is 1 if a team is home, 0 if
a team didn't play, and -1 if a team is a visitor. Anything passed to `adjust()`

in
`formula`

is also put in the data.frame. A `glm`

model is then
run to predict wins or margin of victory.

With this setup, the intercept represents the home-field advantage. Neutral fields can be indicated
using the `neutral()`

function, which sets the intercept to 0.

Note that any weights specified in `players()`

will be ignored.

This is essentially the Bradley-Terry model.

https://en.wikipedia.org/wiki/Bradley

data(tournament) elo.glm(score(points.Home, points.Visitor) ~ team.Home + team.Visitor, data = tournament, subset = points.Home != points.Visitor) #> #> Call: stats::glm(formula = wins.A ~ . - 1, family = family, data = dat.qr, #> weights = wts, subset = NULL, na.action = stats::na.pass) #> #> Coefficients: #> home.field `Athletic Armadillos` `Blundering Baboons` #> 1.0307 1.4289 -0.9637 #> `Cunning Cats` `Defense-less Dogs` `Elegant Emus` #> 0.5377 -1.7413 0.3931 #> `Fabulous Frogs` `Gallivanting Gorillas` #> 0.8489 0.3994 #> #> Degrees of Freedom: 51 Total (i.e. Null); 43 Residual #> Null Deviance: 70.7 #> Residual Deviance: 48.04 AIC: 64.04 elo.glm(mov(points.Home, points.Visitor) ~ team.Home + team.Visitor, data = tournament, family = "gaussian") #> #> Call: stats::glm(formula = wins.A ~ . - 1, family = family, data = dat.qr, #> weights = wts, subset = NULL, na.action = stats::na.pass) #> #> Coefficients: #> home.field `Athletic Armadillos` `Blundering Baboons` #> 2.6964 3.1250 -2.4375 #> `Cunning Cats` `Defense-less Dogs` `Elegant Emus` #> 0.3125 -3.5000 -0.9375 #> `Fabulous Frogs` `Gallivanting Gorillas` #> 0.6875 0.2500 #> #> Degrees of Freedom: 56 Total (i.e. Null); 48 Residual #> Null Deviance: 2161 #> Residual Deviance: 1295 AIC: 352.8