Basketball Data Sciene Basics
Recently I started reading Bsketball Data Sciene…. here are some things.
Load Packages
Using ..
library(BasketballAnalyzeR)
library(ggplot2)Data
BAR <- data(package = "BasketballAnalyzeR")
head(BAR)## $title
## [1] "Data sets"
##
## $header
## NULL
##
## $results
## Package
## [1,] "BasketballAnalyzeR"
## [2,] "BasketballAnalyzeR"
## [3,] "BasketballAnalyzeR"
## [4,] "BasketballAnalyzeR"
## [5,] "BasketballAnalyzeR"
## LibPath Item
## [1,] "/Library/Frameworks/R.framework/Versions/4.0/Resources/library" "Obox"
## [2,] "/Library/Frameworks/R.framework/Versions/4.0/Resources/library" "PbP.BDB"
## [3,] "/Library/Frameworks/R.framework/Versions/4.0/Resources/library" "Pbox"
## [4,] "/Library/Frameworks/R.framework/Versions/4.0/Resources/library" "Tadd"
## [5,] "/Library/Frameworks/R.framework/Versions/4.0/Resources/library" "Tbox"
## Title
## [1,] "Opponents box scores dataset - NBA 2017-2018"
## [2,] "Play-by-play dataset - NBA 2017-2018"
## [3,] "Players box scores dataset - NBA 2017-2018"
## [4,] "Tadd dataset - NBA 2017-2018"
## [5,] "Teams box scores dataset - NBA 2017-2018"
##
## $footer
## NULLThere are 5 data sets:
Obox: Opponents box scores dataset - NBA 2017-2018
Pbp.BDB: Play by Play dataset - NBA 2017-2018
Pbox: Players box scores dataset - NBA 2017-2018
Tadd: Team information such as Conference, Division, final rank, qualification for Playoffs for the NBA 2017-2018 Championship.
Tbox: Team box scores dataset - NBA 2017-2018
Pbp <- PbPmanipulation(PbP.BDB)
head(Pbp)## game_id data_set date a1 a2
## 478 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## 479 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## 480 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## 481 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## 482 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## 483 21700002 2017-2018 Regular Season 10/17/2017 Trevor Ariza Ryan Anderson
## a3 a4 a5 h1 h2
## 478 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## 479 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## 480 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## 481 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## 482 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## 483 Clint Capela James Harden Chris Paul Kevin Durant Draymond Green
## h3 h4 h5 period away_score home_score
## 478 Zaza Pachulia Klay Thompson Stephen Curry 1 0 0
## 479 Zaza Pachulia Klay Thompson Stephen Curry 1 0 0
## 480 Zaza Pachulia Klay Thompson Stephen Curry 1 2 0
## 481 Zaza Pachulia Klay Thompson Stephen Curry 1 2 0
## 482 Zaza Pachulia Klay Thompson Stephen Curry 1 2 0
## 483 Zaza Pachulia Klay Thompson Stephen Curry 1 2 0
## remaining_time elapsed play_length play_id team event_type assist
## 478 0:12:00 0:00:00 0:00:00 2 start of period
## 479 0:12:00 0:00:00 0:00:00 4 HOU jump ball
## 480 0:11:47 0:00:13 0:00:13 7 HOU shot
## 481 0:11:27 0:00:33 0:00:20 8 GSW miss
## 482 0:11:23 0:00:37 0:00:04 9 HOU rebound
## 483 0:11:13 0:00:47 0:00:10 10 HOU miss
## away home block entered left num opponent outof
## 478 NA NA
## 479 Clint Capela Zaza Pachulia NA NA
## 480 NA NA
## 481 NA NA
## 482 NA NA
## 483 NA NA
## player points possession reason result steal type
## 478 NA start of period
## 479 Zaza Pachulia NA Ryan Anderson jump ball
## 480 James Harden 2 made unknown
## 481 Draymond Green 0 missed Jump Shot
## 482 Clint Capela NA rebound defensive
## 483 James Harden 0 missed Running Jump Shot
## shot_distance original_x original_y converted_x converted_y
## 478 NA NA NA NA NA
## 479 NA NA NA NA NA
## 480 2 -10 16 26.0 6.6
## 481 25 9 254 25.9 63.6
## 482 NA NA NA NA NA
## 483 10 46 94 20.4 14.4
## description periodTime totalTime
## 478 0 0
## 479 Jump Ball Pachulia vs. Capela: Tip to Anderson 0 0
## 480 Harden 2' Driving Layup (2 PTS) 13 13
## 481 MISS Green 25' 3PT Jump Shot 33 33
## 482 Capela REBOUND (Off:0 Def:1) 37 37
## 483 MISS Harden 10' Running Jump Shot 47 47
## playlength ShotType oppTeam
## 478 0 <NA>
## 479 0 <NA> GSW
## 480 13 2P GSW
## 481 20 3P HOU
## 482 4 <NA> GSW
## 483 10 2P GSWConcepts (MATH)
Possession and Pace
\[\text{POSS = (P2A + P3A) + 0.44 }\times \text{ FTA - OREB + TOV}\]
\[\text{PACE = 5 }\times\text{ POSS/MIN}\]
Where MIN is the total minutes played by all the players.
Offensive and Defensive Ratings
Pre-possession efficiency is measure as the points scored or allowed per 100 possession, called Offensive (ORtg) and Defensive (DRtg) Ratings respectively:
\[\text{ORtg = PTS}_\text{T}/\text{POSS}_{\text{T}}\]
\[\text{DRtg = PTS}_\text{O}/\text{POSS}_{\text{O}}\]
Where T and O refer to team (being analyzed) and opponent(s).
Four Factors
\(\underline{\text{Efficient Field Goal Percentage}}\)
\[\text{eFG }\%_{\text{ Offense}}=\frac{\text{P2M}_\text{T}+1.5\times\text{ P3M}_\text{T}}{\text{P2A}_\text{T}+\text{P3A}_\text{T}}\]
\[\text{eFG }\%_\text{ Defense}=\frac{\text{P2M}_\text{O}+1.5\times\text{ P3M}_\text{O}}{\text{P2A}_\text{O}+\text{P3A}_\text{O}}\]
\(\underline{\text{Turnovers Per Possession Ratio}}\)
\[\text{TO Ratio}_\text{ Offense}=\frac{\text{TOV}_\text{T}}{\text{POSS}_\text{T}}\]
\[\text{TO Ratio}_\text{ Defense}=\frac{\text{TOV}_\text{O}}{\text{POSS}_\text{O}}\]
\(\underline{\text{Rebounding Percentage}}\)
\[\text{REB }\%_{\text{ Offense}}=\]
\[\text{REB }\%_\text{ Defense}=\]
\(\underline{\text{Free Throw Rate}}\)
\[\text{FT Rate}_{\text{ Offense}}=\]
\[\text{FT Rate}_{\text{ Defense}}=\]
Plots
tm <- c("BOS", "CLE", "GSW", "HOU")
selTeams <- which(Tadd$team %in% tm)
FF.sel <- fourfactors(Tbox[selTeams,], Obox[selTeams,])
plot(FF.sel)
# rearrange
listPlots <- plot(FF.sel)
library(gridExtra)
grid.arrange(grobs=listPlots[1:2], ncol=1)
grid.arrange(grobs=listPlots[3:4], ncol=1)
Bar line Plots
X <- data.frame(Tbox,
PTS.O = Obox$PTS,
TOV.O = Obox$TOV,
CONF = Tadd$Conference)
XW <- subset(X, CONF== "W")
labs <- c("Steals", "Blocks", "Defensive Rebounds")
barline(data = XW,
id = "Team",
bars = c("STL", "BLK", "DREB"),
line = "TOV.O",
order.by = "PTS.O",
labels.bars = labs)
Pbox.HR <- subset(Pbox,
Team=="Phoenix Suns" & MIN>= 500)
Pbox.HR## Team Player GP MIN PTS P2M P2A P2p P3M P3A
## 1918 Phoenix Suns TJ Warren 65 2142 1271 509 973 52.31244 20 90
## 823 Phoenix Suns Dragan Bender 82 2069 531 69 162 42.59259 118 322
## 1523 Phoenix Suns Josh Jackson 77 1959 1012 332 715 46.43357 57 217
## 723 Phoenix Suns Devin Booker 54 1865 1346 308 670 45.97015 147 384
## 2117 Phoenix Suns Tyler Ulis 71 1658 554 172 405 42.46914 42 146
## 2015 Phoenix Suns Troy Daniels 79 1622 703 49 117 41.88034 183 458
## 1623 Phoenix Suns Marquese Chriss 72 1527 556 146 287 50.87108 56 190
## 323 Phoenix Suns Alex Len 69 1395 587 213 375 56.80000 1 3
## 2213 Phoenix Suns Tyson Chandler 46 1151 300 121 187 64.70588 0 0
## 1323 Phoenix Suns Jared Dudley 48 686 152 19 42 45.23810 29 80
## 1722 Phoenix Suns Mike James 32 669 332 95 215 44.18605 26 97
## 923 Phoenix Suns Elfrid Payton 19 551 224 86 184 46.73913 5 25
## P3p FTM FTA FTp OREB DREB AST TOV STL BLK PF PM
## 1918 22.22222 193 255 75.68627 125 208 85 85 65 41 149 -433
## 823 36.64596 39 51 76.47059 40 321 130 112 22 53 166 -469
## 1523 26.26728 177 279 63.44086 95 256 119 146 80 35 218 -529
## 723 38.28125 289 329 87.84195 26 218 253 194 47 14 168 -358
## 2117 28.76712 84 101 83.16832 24 104 311 127 70 7 120 -316
## 2015 39.95633 56 64 87.50000 12 115 48 56 26 5 137 -409
## 1623 29.47368 96 158 60.75949 76 319 83 107 51 70 205 -300
## 323 33.33333 158 231 68.39827 175 343 82 78 27 61 158 -255
## 2213 0.00000 58 93 62.36559 134 284 53 59 16 27 106 -244
## 1323 36.25000 27 35 77.14286 8 86 78 28 24 11 85 2
## 1722 26.80412 64 84 76.19048 12 77 120 48 27 7 45 -30
## 923 20.00000 37 54 68.51852 17 83 118 54 19 6 50 -256barline(data = Pbox.HR, id = "Player",
bars = c("P2p", "P3p", "FTp"),
line = "MIN",
order.by = "PM",
labels.bars = c("2P%", "3P%", "FT%"),
title = "Phoenix Suns")
Radial Plots
Pbox.PG <- subset(Pbox,
Player == "Russell Westbrook" |
Player == "Stephen Curry" |
Player == "Kyrie Irving" |
Player == "Damian Lillard" |
Player == "Kyle Lowry" |
Player == "John Wall" |
Player == "Rajon Rondo" |
Player == "Kemba Walker")
attach(Pbox.PG)
X <- data.frame(P2M, P3M, FTM, REB=OREB+DREB, AST, STL, BLK)/MIN
detach(Pbox.PG)
radialprofile(data = X, title = Pbox.PG$Player, std = F)
radialprofile(data = X, title = Pbox.PG$Player, std = T)
Scatter Plots
Pbox.sel <- subset(Pbox, MIN>=500)
attach(Pbox.sel)
X <- data.frame(AST, TOV, PTS)/MIN
detach(Pbox.sel)
mypal <- colorRampPalette(c("blue", "yellow", "red"))
scatterplot(X, data.var = c("AST", "TOV"), z.var="PTS", labels = 1:nrow(X), palette = mypal)
SAS <- which(Pbox.sel$Team=="San Antonio Spurs")
scatterplot(X, data.var = c("AST", "TOV"), z.var="PTS", labels = Pbox.sel$Player, palette = mypal, subset = SAS)
scatterplot(X, data.var = c("AST", "TOV"), z.var="PTS", labels = Pbox.sel$Player, palette = mypal, subset = SAS, zoom = c(0.2, 0.325, 0.05, 0.10))
Bubble Plots
attach(Tbox)
X <- data.frame(T = Team, P2p, P3p, FTp, AS = P2A + P3A + FTA)
detach(Tbox)
labs <- c("2-Point shots (% made)",
"3-Point shots (% made)",
"Free throws (% made)",
"Total shots attempted")
bubbleplot(X, id = "T", x = "P2p", y = "P3p", col = "FTp",
size = "AS", labels = labs)
Pbox.CC.GSW <- subset(Pbox,
Team == "Golden State Warriors" | Team == "Cleveland Cavaliers" & MIN >= 500)
attach(Pbox.CC.GSW)
X <- data.frame(ID = Player,
Team,
V1 = DREB/MIN,
V2 = STL/MIN,
V3 = BLK/MIN,
V4 = MIN)
detach(Pbox.CC.GSW)
labs <- c("Defensive Rebounds",
"Steals",
"Blocks",
"Total minutes played")
bubbleplot(X, id = "ID",
x = "V1", y = "V2", col = "V3", size = "V4",
text.col = "Team", labels = labs,
title = "CC and GSW during the regular season",
text.legend = TRUE,
text.size = 3.5,
scale = FALSE)