The goal of nflseedR is to allow NFL modelers to simulate NFL seasons using their models, and taking off their plate the work of tracking the schedule, navigating the complex rules for division ranking, playoff seeding, and draft order. This can also aid in sports betting, such as betting on futures or win totals.
The package can run thousands of Monte Carlo style simulations of the NFL regular season, based on a model you input. Within each simulated season, it will calculate the division standings and playoff seedings for you. It will also the generate the playoff games and simulate these as well, and calculate the order for next year’s NFL draft. These can be used to examine the probability of team making the playoffs or winning the Super Bowl, based on your model.
The season simulations will take all completed games into account already, and only simulate from there forward, including if run during the playoffs. It can also be run as a fresh season, wiping away results and simulating from scratch.
The season simulation code for nflseedR was developed by Lee Sharpe (@LeeSharpeNFL) and building it as package was developed by Sebastian Carl (@mrcaseb). New high efficient standings code was developed by Sebastian Carl (@mrcaseb)
The easiest way to get nflseedR is to install it from CRAN with:
install.packages("nflseedR")To get a bug fix or to use a feature from the development version, you can install the development version of nflseedR either from GitHub with
if (!require("pak")) install.packages("pak")
pak::pak("nflverse/nflseedR")or prebuilt from the development repo with:
install.packages("nflseedR", repos = c("https://nflverse.r-universe.dev", getOption("repos")))With nflseedR it’s possible to simulate complete seasons or use it’s seeding functions in custom simulations.