add.home.bias: Add the effect of home-team bias to population picks

In elishayer/mRchmadness: Numerical Tools for Filling Out an NCAA Basketball Tournament Bracket

Add the effect of home-team bias to population picks

Description

Add the effect of home-team bias to population picks

Usage

add.home.bias(teams, league = c("men", "women"), year = 2018)

Arguments

teams	character vector of names of teams to whom home-team bias is to be applied (must match name column of pred.pop.[league].[year])
league	which league: "men" (default) or "women", for prob.source.
year	year of tournament

Details

This function implements a rule of thumb gleamed from an article published at CBS Sports on March 10, 2016, by Brad Null of Bracket Voodoo, titled "Homer bias is real and it will derail your March Madness bracket". The sparse-detail results published there are consistent with adding a constant +3/4 to the log-odds of the conditional probability of the "home" team winning in each round. For more details, see https://github.com/elishayer/mRchmadness/issues/13. Null cautions in his piece that "schools with smaller undergraduate enrollment (Duke, Xavier, Villanova) tend to have a smaller regional footprint", so only use this adjustment if you have a good reason to think the picks in your pool would be biased (for example a pool of UNC alumni).

Value

a 64-by-6 matrix matching population pick matrix from year, except that specified teams benefit from greater advancement probability attributable to home-team bias. Rows correspond to teams, and columns correspond to rounds, with the values giving the probability of each team winning in each round.

Author(s)

sspowers

Examples

add.home.bias(teams = 'UNC', league = 'men', year = 2018)

elishayer/mRchmadness documentation built on March 27, 2024, 2:11 p.m.