README.md

In choosunsick/kbodatamining: Kbo Data Mining and Analysis with R

kbodatamining

Package Introduction

The goal of kbodatamining package is analizing and visualizing korean baseball data(KBO data). This package is intended to make use of KBO data collected from Python to R.

Package Installation

You can install package kbodatamining from github with the following command.

# install.packages("devtools")
devtools::install_github("choosunsick/kbodatamining")

Package Functionality

There are two primary functions in this package. The first is to make the records of players box-score. for example:

library(kbodatamining)

# batter example
batter_boxscore(data = hanhwa_batter_2018,name = "이용규",yearly = 2018)
#>       name period team   g away home  pa  ab   h  r rbi  1B 2B 3B hr  tb
#> 788 이용규   2018 한화 134   65   69 575 491 144 82  36 128 14  1  1 163
#>     bb hbp ibb so gidp sh sf   avg   obp   slg   ops babip  id
#> 788 59  12   0 62    8  8  5 0.293 0.379 0.332 0.711 0.328 788
# pitcher example
pitcher_boxscore(data = hanhwa_pitcher_2018,name = "정우람",yearly = 2018)
#>     name period team  g away home win lose draw  wpct sv hld ip  r er  k
#> 1 정우람   2018 한화 55   24   31   5    3    0 0.625  0  35 53 22 20 56
#>   bb&hbp pit tbf ha hra   era   p_ip   k_9  hr_9   oba  id cg cgs
#> 1     14 888 220 49   6 3.396 16.755 9.509 1.019 0.245 955  0   0
# team hitting example
team_batter_boxscore(data = hanhwa_batter_2018,teamname = "한화",yearly = 2018)
#>   period team   g away home   pa   ab    h   r rbi  1B  2B 3B  hr   tb  bb
#> 1   2018 한화 144   72   72 5539 4972 1369 729 668 955 249 14 151 2099 420
#>   hbp ibb   so gidp sh sf   avg   obp   slg   ops babip
#> 1  88  17 1098  113 30 28 0.275 0.341 0.422 0.763 0.325
# team pitching example
team_pitcher_boxscore(data = hanhwa_pitcher_2018,teamname = "한화",yearly = 2018)
#>   period team   g away home win lose draw      wpct sv hld   ip   r  er
#> 1   2018 한화 144   72   72  77   67    0 0.5347222 62  37 1274 761 701
#>      k bb&hbp   pit  tbf   ha hra   era   p_ip  k_9  hr_9   oba cg cgs
#> 1 1124    561 22342 5637 1394 164 4.952 17.537 7.94 1.159 0.279  0   0

The second is to visualize monthly or seasonal comparisons for each player. In addition, these two functions can be applied to each team. for example:

library(kbodatamining)
theme_set(theme_gray(base_family = "AppleGothic"))
# comparing batters
oneseason_batter_compare(data = hanhwa_batter_2018,playername1 = "이용규",playername2 = "정근우",yearly=2018)

# comparing pitchers
oneseason_pitcher_compare(data = hanhwa_pitcher_2018,playername1 = "정우람",playername2 = "송은범",yearly=2018)

# comparing batters of team
compare_batter_statistic_teams(hanhwa_batter_2018,"한화","LG",yearly=2018)

# comparing pitchers of team
compare_pitcher_statistic_teams(hanhwa_pitcher_2018,"한화","LG",yearly=2018)

You can find a detailed explanation of this function in the ‘plotting example’ document of the vignette folder.

Notes on data

How to collect KBO data from Python can be found at LINK. For reference, the sample data in the package includes the data of 2018 season Hanhwa team. This sample data is divided into ‘hanhwa_batter_2018’, ‘hanhwa_pitcher_2018’. You can check the package’s functions using sample data.

Json data processing (json data 처리과정)

If you see the link above and get a json file for the KBO game through Python, you will have to process the json file to use it in R. So I’ll show you an example of loading a json file from R and making it to DATAFRAME. An example of the process will be shown through the data in the json_sample folder in this github. for example:

library(jsonlite)
library(kbodatamining)
# read json file
sample_set <- fromJSON("./json_sample/Hanhwa_normalseason_2018.json")
# make batter data dataframe 
# If you want pitch data, use json2pitcherdf instead of json2batterdf
hanhwa_batter_2018 <- do.call(rbind,lapply(names(sample_set),FUN = function(x)json2batterdf(sample_set,x)))
# organizing column order & column name
hanhwa_batter_2018 <- hanhwa_batter_2018[,c(18:21,11,16:17,1:9,24,23,22,13,12,10,15)]
colnames(hanhwa_batter_2018) <- c('date','away','home','doubleheader','name','team','position',
                                  'one','two','three','four','five','six','seven','eight','nine',
                                  'ten','eleven','twelve','ab','h','r','rbi')
head(hanhwa_batter_2018)
#>         date away home doubleheader   name team position  one  two three
#> 1 2018-03-24 한화 넥센            0 이용규 한화       중 2000 7003     0
#> 2 2018-03-24 한화 넥센            0 양성우 한화       좌 7006    0  7107
#> 3 2018-03-24 한화 넥센            0 송광민 한화       一 1017    0  1019
#> 4 2018-03-24 한화 넥센            0 김태균 한화       지 1011    0  1017
#> 5 2018-03-24 한화 넥센            0 이동훈 한화       주    0    0     0
#> 6 2018-03-24 한화 넥센            0 하주석 한화       유 1017    0  7005
#>   four five  six seven eight nine ten eleven twelve ab h r rbi
#> 1    0 7405    0  1019     0 2000   0      0      0  5 1 0   1
#> 2    0 1017    0  7002     0 7101   0      0      0  5 1 0   0
#> 3    0    0 2000     0  7003    0   0      0      0  4 2 1   0
#> 4    0    0 2000     0  1017    0   0      0      0  4 3 0   0
#> 5    0    0    0     0     0    0   0      0      0  0 0 0   0
#> 6    0    0 7003     0  7100    0   0      0      0  4 1 0   1

In addition, you need to create an id column of the players.

library("kbodatamining")
# read id list 
batter_id  <- read.csv("./player_id_list/batter_id_list.csv",stringsAsFactors = F)
# extract the appropriate id list for your data
batter_id_2018 <- batter_id[substr(batter_id$date,1,4)==2018,]
batter_id_hanhwa_2018 <- batter_id_2018[batter_id_2018$away=="한화"|batter_id_2018$home=="한화",]
# make id column 
hanhwa_batter_2018 <- get_player_id(hanhwa_batter_2018,batter_id_hanhwa_2018)
head(hanhwa_batter_2018)
#>            date away home doubleheader   name team position  one two three
#> 1350 2018-06-03 한화 롯데            0 강경학 한화       三    0   0     0
#> 1374 2018-06-05 한화   LG            0 강경학 한화       유    0   0     0
#> 1393 2018-06-06 한화   LG            0 강경학 한화       타    0   0     0
#> 1416 2018-06-07 한화   LG            0 강경학 한화       주    0   0     0
#> 1447 2018-06-08   SK 한화            0 강경학 한화       유 1300   0  1110
#> 1472 2018-06-09   SK 한화            0 강경학 한화       유 2000   0  1019
#>      four five  six seven eight nine ten eleven twelve ab h r rbi id
#> 1350    0    0    0     0     0    0   0      0      0  0 0 0   0  5
#> 1374    0    0    0  1019     0 1019   0      0      0  2 2 0   1  5
#> 1393    0    0    0     0     0    0   0      0      0  0 0 0   0  5
#> 1416    0    0    0     0     0    0   0      0      0  0 0 0   0  5
#> 1447 2000    0 1017     0  2000    0   0      0      0  5 3 2   2  5
#> 1472    0 7005    0     0  2000    0   0      0      0  4 1 0   0  5

Repeat the above process with your json file, changing only the year and team name.

choosunsick/kbodatamining documentation built on May 18, 2019, 1:30 p.m.