cfb4th

This is the package that powers the A.I. Sports College Football 4th Down Model discussed in this UteZone article. The model is heavily based on the work and code produced by Ben Baldwin for his 4th Down Calculator for the NFL

Data freshness and pipeline status for every SportsDataverse dataset: sportsdataverse.org/status.

Installation

You can install the released version of cfb4th from CRAN with:

install.packages("cfb4th")

And the development version from GitHub with:

# install.packages("pak")
# pak::pak("sportsdataverse/cfb4th")
install.packages("cfb4th", repos = c('https://sportsdataverse.r-universe.dev', 'https://cloud.r-project.org'))

The two xgboost models (fourth-down conversion and win probability) are not bundled with the package: each is downloaded the first time it is needed and cached under tools::R_user_dir("cfb4th", "cache"), so the first call to add_4th_probs() needs network access. Clear the cache to force a fresh download with cfb4th_clear_cache().

Features

• The go for it model gives probabilities for possibilities of yards gained and includes the possibility of earning a first down via defensive penalty
• The punt model includes the possibility for getting blocked, returned for a touchdown, or fumbled on the return
• The field goal model is a simple model of field goal % by distance and roof type

Current limitations

There are some edge cases that are not accounted for. These should only make a marginal difference to the recommendations as they are largely edge cases (e.g. the possibility for a field goal to be blocked and returned).

• The go for it model does not allow for the possibility of a turnover return. However, long returns are extremely rare: For example, in 2018 and 2019 in the NFL there were only four defensive touchdowns on plays where teams went for fourth downs out of 1,236 plays, and all of these happened when the game was well in hand for the other team. Additionally, it assumes that a touchdown is worth 7 points and doesn’t account for go-for-2 situations (Future Work)
• The punt model doesn’t account for the punter or returner, ignores penalties on returns and ignores the potential for blocked punts to be returned for touchdowns
• The field goal model doesn’t account for who the kicker is, what the weather is (only relevant for outdoor games), or the possibility of a kick being blocked and returned for a touchdown

Get Started

To get started with cfb4th please see this article.

• The logic of the shiny app

• The code that powers the bot

• The bot can be found on twitter here.

Follow the SportsDataverse (@SportsDataverse) on X and star this repo

Our Authors

• Jared Lee

• Sebastian Carl
  

• Ben Baldwin
  

Cheat sheet

A printable one-page reference covering cfbplotR, cfb4th and cfbseedR — the function families, the loaders, and what each one returns.

📄 Download the cfb4th cheat sheet (PDF)

Free to download, print and hand out; light and dark, US Letter landscape. Every SportsDataverse package has one — browse them all at sportsdataverse.org/cheatsheets.

Citations

To cite the cfb4th R package in publications, use:

BibTeX Citation

@misc{lee_et_al_2021_cfb4th,
  author = {Jared Lee and Ben Baldwin and Sebastian Carl and Saiem Gilani},
  title = {cfb4th: The SportsDataverse's R Package for College Football 4th Down Modeling.},
  url = {https://cfb4th.sportsdataverse.org/},
  year = {2026}
}