Package 'ffopportunity'

Build EP

Description

This function builds Expected Fantasy Points predictions by downloading the xgboost models and play-by-play data, applying the model, and summarizing to player level.

Usage

ep_build(season = nflreadr::most_recent_season(), version = "latest")

Arguments

season	a numeric vector of seasons that defaults to most recent season. Must be later than 2006.
version	an EP model version - one of "latest" (default) or "v1.0.0" (these are currently identical)

Value

a list containing three dataframes: ep_weekly provides a game-level summary by player, ep_pbp_pass provides EP data on pass plays, and ep_pbp_rush provides EP data on rush plays.

See Also

Other main: ep_load()

Examples

try({ # prevents cran-related errors
  ep_build(season = 2021)
})

---

Model versioning

Description

This function checks the cache for a previously downloaded model and then (optionally) tries to download the model from GitHub release.

Usage

ep_cache_models(
  version = c("latest", "v1.0.0"),
  force = FALSE,
  ask = interactive()
)

Arguments

version	one of "latest" or "v1.0.0" - currently these refer to the same thing
force	TRUE or FALSE - forces download regardless of currently existing
ask	TRUE or FALSE - ask before downloading - force will skip this.

Value

a status message after attempting to download the model.

---

Load Expected Points data

Description

This function downloads precomputed expected points data from the ffopportunity automated releases.

Usage

ep_load(
  season = nflreadr::most_recent_season(),
  type = c("weekly", "pbp_pass", "pbp_rush"),
  version = c("latest", "v1.0.0")
)

Arguments

season	A numeric vector of four digit years associated with given NFL seasons - defaults to latest season.
type	Data type - one of "weekly", "pbp_pass", or "pbp_rush"
version	EP model version: one of "latest" (default) or "v1.0.0" - these are currently identical.

Value

a dataframe identical to what would be returned by ffopportunity::ep_build() for a given season.

See Also

Other main: ep_build()

Examples

try({
  ep_load() %>% head(10)
  ep_load(2020:2021) %>% head(10)
  ep_load(2021, type = "pbp_pass") %>% head(10)
  ep_load(2006, type = "pbp_rush", version = "v1.0.0") %>% head(10)
})

---

Predict EP

Description

This function runs the prediction functions over preprocessed data.

Usage

ep_predict(preprocessed_pbp, version = c("latest", "v1.0.0"))

Arguments

preprocessed_pbp	list with dataframes created by ep_preprocess
version	ep model version - available is "latest" or "v1.0.0" (these are currently the same thing)

Value

a dataframe with the expected fields added

Examples

try({
  preprocessed <- readRDS(system.file("ep_preprocessed.rds",package = "ffopportunity"))
  # this file is equivalent to nflreadr::load_pbp(2021) %>% head(1000) %>% ep_preprocess()
  ep_predict(preprocessed)
  })

---

Preprocess Data

Description

This function performs pre-processing steps to make expected points predictions on nflreadr data

Usage

ep_preprocess(pbp)

Arguments

pbp	pbp dataframe from nflreadr::load_pbp()

Value

a list of two dataframes (one for passes and one for rushes) of nflreadr data with the expectedpoints columns transformed for prediction

See Also

vignette("basic") for example usage

Examples

try({ # catch failures for CRAN purposes
  pbp_download <- readRDS(system.file("pbp_download.rds",package = "ffopportunity"))
  # this file is equivalent to nflreadr::load_pbp(2021) %>% head(1000)
  ep_preprocess(pbp_download)
  })

---

Summarize EP

Description

This function summarizes the EP data up to the game level

Usage

ep_summarize(
  predicted_pbp,
  stat_type = c("all", "expected_points", "team_stats")
)

Arguments

predicted_pbp	list with dataframes created by ep_predict
stat_type	options to limit the columns returned by ep_summarize - available options are "all", "expected_points", and "team stats"

Value

a dataframe with the expected points fields added

See Also

vignette("basic") for example usage

Examples

try({
  predicted <- readRDS(system.file("ep_predicted.rds",package = "ffopportunity"))
  # equivalent to nflreadr::load_pbp(2021) %>% head(100) %>% ep_preprocess() %>% ep_predict()
  ep_summarize(predicted)
})

---