gbbo

A comprehensive, tidy dataset and companion R package for the hit television series The Great British Bake Off (GBBO), broadcast as The Great British Baking Show in North America.

Installation

Install the development version of gbbo directly from GitHub:

# install.packages("remotes")
remotes::install_github("nicholaswilde/the-great-british-bake-off/r")

Or install via R-universe:

install.packages("gbbo", repos = "https://nicholaswilde.r-universe.dev")

Datasets Included

The package exports 4 clean, relational data frames:

Dataset	Description
seasons	Season broadcast years, winners, hosts, judges, networks, and streaming links
bakers	Baker demographics, ages, hometowns, occupations, and overall finish placements
episodes	Episode air dates, themes, and challenge duration time limits
performances	Challenge bakes, technical ranks, Hollywood Handshakes, and results

Quickstart

library(gbbo)
library(dplyr)
library(ggplot2)

# Inspect seasons overview
head(seasons)

# Hollywood Handshake distribution across series
performances |>
  filter(hollywood_handshake) |>
  count(episode_id, sort = TRUE)

# Star Baker frequencies
performances |>
  filter(result == "STAR_BAKER") |>
  count(baker_id, sort = TRUE)

Direct Parquet / Arrow Usage

For users who prefer high-performance columnar reads without loading package .rda objects, read the pre-compiled Parquet tables directly using the {arrow} package:

library(arrow)

seasons <- read_parquet("https://github.com/nicholaswilde/the-great-british-bake-off/raw/main/data/processed/seasons.parquet")
bakers <- read_parquet("https://github.com/nicholaswilde/the-great-british-bake-off/raw/main/data/processed/bakers.parquet")

License

This package is licensed under the Apache License 2.0.