Skip to content

A package written for R focused on baseball analysis. Currently in development.

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md
Notifications You must be signed in to change notification settings

BillPetti/baseballr

Repository files navigation

baseballr

CRAN version CRAN downloads Version-Number R-CMD-check Lifecycle:maturing Contributors

baseballr is a package written for R focused on baseball analysis. It includes functions for scraping various data from websites, such as FanGraphs.com, Baseball-Reference.com, and baseballsavant.mlb.com. It also includes functions for calculating metrics, such as wOBA, FIP, and team-level consistency over custom time frames.

You can read more about some of the functions and how to use them at its official site as well as this Hardball Times article.

Installation

You can install the CRAN version of baseballr with:

install.packages("baseballr")

You can install the released version of baseballr from GitHub with:

# You can install using the pacman package using the following code:
if (!requireNamespace('pacman', quietly = TRUE)){
  install.packages('pacman')
}
pacman::p_load_current_gh("BillPetti/baseballr")
# Alternatively, using the devtools package:
if (!requireNamespace('devtools', quietly = TRUE)){
  install.packages('devtools')
}
devtools::install_github(repo = "BillPetti/baseballr")

For experimental functions in development, you can install the development branch:

# install.packages("devtools")
devtools::install_github("BillPetti/baseballr", ref = "development_branch")

Functionality

The package consists of two main sets of functions: data acquisition and metric calculation.

For example, if you want to see the standings for a specific MLB division on a given date, you can use the bref_standings_on_date() function. Just pass the year, month, day, and division you want:

library(baseballr)
library(dplyr)
bref_standings_on_date("2015-08-01", "NL East", from = FALSE)
## ── MLB Standings on Date data from baseball-reference.com ─── baseballr 1.5.0 ──

## ℹ Data updated: 2023-12-25 02:24:44 EST

## # A tibble: 5 × 8
##   Tm        W     L `W-L%` GB       RS    RA `pythW-L%`
##   <chr> <int> <int>  <dbl> <chr> <int> <int>      <dbl>
## 1 WSN      54    48  0.529 --      422   391      0.535
## 2 NYM      54    50  0.519 1.0     368   373      0.494
## 3 ATL      46    58  0.442 9.0     379   449      0.423
## 4 MIA      42    62  0.404 13.0    370   408      0.455
## 5 PHI      41    64  0.39  14.5    386   511      0.374

Right now the function works as far as back as 1994, which is when both leagues split into three divisions.

You can also pull data for all hitters over a specific date range. Here are the results for all hitters from August 1st through October 3rd during the 2015 season:

data <- bref_daily_batter("2015-08-01", "2015-10-03") 
data %>%
  dplyr::glimpse()
## Rows: 764
## Columns: 30
## $ bbref_id <chr> "machama01", "duffyma01", "altuvjo01", "eatonad02", "choosh01…
## $ season   <int> 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2…
## $ Name     <chr> "Manny Machado", "Matt Duffy", "José Altuve", "Adam Eaton", "…
## $ Age      <dbl> 22, 24, 25, 26, 32, 21, 27, 28, 36, 28, 29, 29, 27, 29, 27, 2…
## $ Level    <chr> "Maj-AL", "Maj-NL", "Maj-AL", "Maj-AL", "Maj-AL", "Maj-AL", "…
## $ Team     <chr> "Baltimore", "San Francisco", "Houston", "Chicago", "Texas", …
## $ G        <dbl> 59, 59, 57, 58, 58, 58, 59, 58, 59, 57, 55, 57, 57, 58, 56, 5…
## $ PA       <dbl> 266, 264, 262, 262, 260, 259, 259, 258, 257, 257, 255, 255, 2…
## $ AB       <dbl> 237, 248, 244, 230, 211, 224, 239, 235, 231, 233, 213, 218, 2…
## $ R        <dbl> 36, 33, 30, 37, 48, 35, 32, 29, 37, 27, 50, 37, 36, 25, 38, 4…
## $ H        <dbl> 66, 71, 81, 74, 71, 79, 54, 66, 75, 48, 65, 56, 61, 51, 78, 5…
## $ X1B      <dbl> 43, 54, 53, 56, 47, 51, 34, 37, 48, 30, 34, 32, 35, 33, 66, 2…
## $ X2B      <dbl> 10, 12, 19, 12, 14, 17, 6, 17, 16, 11, 13, 13, 15, 10, 7, 13,…
## $ X3B      <dbl> 0, 2, 3, 1, 1, 4, 1, 0, 2, 1, 2, 4, 0, 1, 3, 0, 4, 0, 1, 1, 0…
## $ HR       <dbl> 13, 3, 6, 5, 9, 7, 13, 12, 9, 6, 16, 7, 11, 7, 2, 20, 9, 8, 8…
## $ RBI      <dbl> 32, 30, 18, 31, 34, 32, 27, 40, 53, 21, 50, 19, 31, 39, 23, 4…
## $ BB       <dbl> 26, 15, 10, 23, 39, 18, 16, 17, 21, 21, 34, 33, 21, 39, 12, 3…
## $ IBB      <dbl> 1, 0, 1, 1, 1, 0, 0, 6, 1, 1, 0, 1, 1, 5, 0, 4, 3, 3, 7, 2, 2…
## $ uBB      <dbl> 25, 15, 9, 22, 38, 18, 16, 11, 20, 20, 34, 32, 20, 34, 12, 35…
## $ SO       <dbl> 42, 35, 28, 55, 51, 38, 68, 56, 29, 53, 46, 62, 41, 48, 27, 7…
## $ HBP      <dbl> 2, 0, 4, 5, 8, 1, 3, 5, 1, 1, 2, 3, 3, 1, 1, 6, 1, 3, 4, 1, 0…
## $ SH       <dbl> 0, 0, 1, 2, 1, 11, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 0, 0, …
## $ SF       <dbl> 1, 1, 3, 2, 1, 5, 1, 1, 4, 2, 5, 1, 2, 2, 3, 0, 3, 2, 3, 4, 3…
## $ GDP      <dbl> 5, 9, 6, 1, 1, 4, 2, 2, 9, 7, 5, 1, 4, 8, 1, 2, 3, 10, 5, 4, …
## $ SB       <dbl> 6, 8, 11, 9, 2, 10, 0, 0, 0, 3, 3, 4, 5, 4, 24, 2, 1, 0, 6, 0…
## $ CS       <dbl> 4, 0, 4, 4, 0, 2, 0, 0, 0, 1, 0, 1, 3, 2, 7, 2, 3, 0, 2, 0, 0…
## $ BA       <dbl> 0.279, 0.286, 0.332, 0.322, 0.337, 0.353, 0.226, 0.281, 0.325…
## $ OBP      <dbl> 0.353, 0.326, 0.364, 0.392, 0.456, 0.395, 0.282, 0.341, 0.377…
## $ SLG      <dbl> 0.485, 0.387, 0.508, 0.448, 0.540, 0.558, 0.423, 0.506, 0.528…
## $ OPS      <dbl> 0.839, 0.713, 0.872, 0.840, 0.996, 0.953, 0.705, 0.848, 0.906…

In terms of metric calculation, the package allows the user to calculate the consistency of team scoring and run prevention for any year using team_consistency():

team_consistency(2015)
## # A tibble: 30 × 5
##    Team  Con_R Con_RA Con_R_Ptile Con_RA_Ptile
##    <chr> <dbl>  <dbl>       <dbl>        <dbl>
##  1 ARI    0.37   0.36          17           15
##  2 ATL    0.41   0.4           88           63
##  3 BAL    0.4    0.38          70           42
##  4 BOS    0.39   0.4           52           63
##  5 CHC    0.38   0.41          30           85
##  6 CHW    0.39   0.4           52           63
##  7 CIN    0.41   0.36          88           15
##  8 CLE    0.41   0.4           88           63
##  9 COL    0.35   0.34           7            3
## 10 DET    0.39   0.38          52           42
## # ℹ 20 more rows

You can also calculate wOBA per plate appearance and wOBA on contact for any set of data over any date range, provided you have the data available.

Simply pass the proper data frame to woba_plus:

data %>%
  dplyr::filter(PA > 200) %>%
  woba_plus %>%
  dplyr::arrange(desc(wOBA)) %>%
  dplyr::select(Name, Team, season, PA, wOBA, wOBA_CON) %>%
  dplyr::glimpse()
## Rows: 117
## Columns: 6
## $ Name     <chr> "Edwin Encarnación", "Bryce Harper", "David Ortiz", "Joey Vot…
## $ Team     <chr> "Toronto", "Washington", "Boston", "Cincinnati", "Baltimore",…
## $ season   <int> 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2…
## $ PA       <dbl> 216, 248, 213, 251, 253, 260, 245, 255, 223, 241, 223, 259, 2…
## $ wOBA     <dbl> 0.490, 0.450, 0.449, 0.445, 0.434, 0.430, 0.430, 0.422, 0.410…
## $ wOBA_CON <dbl> 0.555, 0.529, 0.541, 0.543, 0.617, 0.495, 0.481, 0.494, 0.459…

You can also generate these wOBA-based stats, as well as FIP, for pitchers using the fip_plus() function:

bref_daily_pitcher("2015-04-05", "2015-04-30") %>% 
  fip_plus() %>% 
  dplyr::select(season, Name, IP, ERA, SO, uBB, HBP, HR, FIP, wOBA_against, wOBA_CON_against) %>%
  dplyr::arrange(dplyr::desc(IP)) %>% 
  head(10)
## ── MLB Daily Pitcher data from baseball-reference.com ─────── baseballr 1.5.0 ──

## ℹ Data updated: 2023-12-25 02:27:52 EST

## # A tibble: 10 × 11
##    season Name               IP   ERA    SO   uBB   HBP    HR   FIP wOBA_against
##     <int> <chr>           <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>        <dbl>
##  1   2015 Johnny Cueto     37    1.95    38     4     2     3  2.62        0.21 
##  2   2015 Dallas Keuchel   37    0.73    22    11     0     0  2.84        0.169
##  3   2015 Sonny Gray       36.1  1.98    25     6     1     1  2.69        0.218
##  4   2015 Mike Leake       35.2  3.03    25     7     0     5  4.16        0.24 
##  5   2015 Félix Hernández  34.2  1.82    36     6     3     1  2.2         0.225
##  6   2015 Corey Kluber     34    4.24    36     5     2     2  2.4         0.295
##  7   2015 Jake Odorizzi    33.2  2.41    26     8     1     0  2.38        0.213
##  8   2015 Josh Collmenter  32.2  2.76    16     3     0     1  2.82        0.29 
##  9   2015 Bartolo Colón    32.2  3.31    25     1     0     4  3.29        0.28 
## 10   2015 Zack Greinke     32.2  1.93    27     7     1     2  3.01        0.24 
## # ℹ 1 more variable: wOBA_CON_against <dbl>

Issues

Please leave any suggestions or bugs in the Issues section.

Pull Requests

Pull request are welcome, but I cannot guarantee that they will be accepted or accepted quickly. Please make all pull requests to the development branch for review.

Breaking Changes

Full News on Releases

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

GitHub stars

Our Authors

  • Bill Petti (@BillPetti)

    @BillPetti

  • Saiem Gilani (@saiemgilani)

    @saiemgilani

Our Contributors (they’re awesome)

  • Ben Baumer (@BaumerBen)

    @beanumber

  • Ben Dilday (@BenDilday)

    @bdilday

  • Robert Frey (@RobertFrey40)

    @robert-frey

  • Camden Kay (@k_camden)

    @camdenk

Citations

To cite the baseballr R package in publications, use:

BibTex Citation

@misc{petti_gilani_2021,
  author = {Bill Petti and Saiem Gilani},
  title = {baseballr: The SportsDataverse's R Package for Baseball Data.},
  url = {https://billpetti.github.io/baseballr/},
  year = {2021}
}