Final Project

Final_Project

Celtics 2021-2022

Author

Daniel Manning

Published

February 5, 2023

Code

library(tidyverse)
library(ggplot2)
library(here)

knitr::opts_chunk$set(echo = TRUE, warning=FALSE, message=FALSE)

Description of Dataset

The NBA Player Stats dataset was sourced from Basketball-Reference.com, a reputable source for broad categories of professional basketball statistics. These statistics are collected by the leagues statisticians through the season and are regarded as very reliable. The dataset includes statistics for every player in the NBA (2374) from the 1997-1998 season to the 2021-2022 season, across 38 teams, so it contains a massive amount of information. This information is broken down into 14573 player-season instances (rows) described by 31 variables (columns). The variable keys and descriptions are listed below for explanation along with their type and cases.

Rk: index Player: player name (2374) Pos: position (string) (5 different main cases, 11 combination cases) Tm: team (string) (38) G: games played (numeric) GS: games started (numeric) MP: minutes played per game (numeric) FG: field goals made per game (numeric) FGA: field goals attempted per game (numeric) FG%: field goal percentage (numeric) 3P: three-pointers made per game (numeric) 3PA: three-pointers attempted per game (numeric) 3P%: three-point percentage (numeric) 2P: two-pointers made per game (numeric) 2PA: two-pointers attempted per game (numeric) 2P%: two-point percentage (numeric) eFG%: effective field goal percentage (numeric) FT: free throws made per game (numeric) FTA: free throws attempted per game (numeric) FT%: free throw percentage (numeric) ORB: offensive rebounds per game (numeric) DRB: defensive rebounds per game (numeric) TRB: total rebounds per game (numeric) AST: assists per game (numeric) STL: steals per game (numeric) BLK: blocks per game (numeric) TOV: turnovers per game (numeric) PF: personal fouls per game (numeric) PTS: points per game (numeric) Year: NBA season year (ordinal) (25)

Load Dataset, Obtain Dimensions, and Unique Cases

Code

NBA<- here("posts","_data","NBA_Player_Stats.csv")%>%
  read_csv()

dim(NBA)

[1] 14573    31

Code

n_distinct(NBA$Player)

[1] 2374

Code

n_distinct(NBA$Pos)

[1] 16

Code

n_distinct(NBA$Tm)

[1] 38

Code

n_distinct(NBA$Year)

[1] 25

Selecting subsets of the data

I chose to focus on data from the Celtics’ and Warriors’ 2021-2022 season in order to compare the two teams who appeared in the NBA finals across some of the major statistics included in the NBA_player_Stats dataset.

Code

last_season <- NBA[NBA$Year == '2021-2022',]
Celtics <- last_season[last_season$Tm == 'BOS',]
Warriors <- last_season[last_season$Tm == 'GSW',]

Change Names of Columns into Recognizable Values/Statistics

Next, I recoded the variable keys into the categories/statistics that they describe, for ease of interpretation.

Code

colnames(Celtics) <- c("Index", "Player", "Position", "Age", "Team", "Games Played", "Games Started", "Minutes Played", "Field Goals Made", "Field Goals Attempted", "Field Goal Percentage", "Three Point Made", "Three Point Attempted", "Three Point Percentage", "Two Point Made", "Two Point Attempted", "Two Point Percentage", "Effective Field Goal Percentage", "Free Throws Made", "Free Throws Attempted", "Free Throw Percentage", "Offensive Rebounds", "Defensive Rebounds", "Total Rebounds", "Assists", "Steals", "Blocks", "Turnovers", "Personal Fouls", "Points", "Season")

colnames(Warriors) <- c("Index", "Player", "Position", "Age", "Team", "Games Played", "Games Started", "Minutes Played", "Field Goals Made", "Field Goals Attempted", "Field Goal Percentage", "Three Point Made", "Three Point Attempted", "Three Point Percentage", "Two Point Made", "Two Point Attempted", "Two Point Percentage", "Effective Field Goal Percentage", "Free Throws Made", "Free Throws Attempted", "Free Throw Percentage", "Offensive Rebounds", "Defensive Rebounds", "Total Rebounds", "Assists", "Steals", "Blocks", "Turnovers", "Personal Fouls", "Points", "Season")

Main Research Question and Subsetting of Data

My research question focuses the differences between the Celtics and Warriors starters and bench player statistics. Analysts often focus on the performance of these two groups when discussing team match-ups, as teams often differ in the contribution of their starters and bench, in terms of efficiency, minutes played, etc. Teams with greater contribution from their bench players are said to have greater depth, which can allow a team to withstand injuries throughout the season, for top players to perform optimally, and for coaches to adjust personnel for specific match-ups. In order to avoid players with limited games played to bias the statistics collected, I removed players who played less than ten games. I then ordered the players from each team by descending order of games started, selected the first five as the starting unit, and the remaining players as the bench unit.

Code

Celtics <- Celtics[(Celtics$"Games Played" >= 10),]
Celtics_by_starts <- Celtics[order(Celtics$"Games Started",decreasing=TRUE),]
Celtics_starters <- Celtics_by_starts[1:5,]
Celtics_bench <- Celtics_by_starts[6:20,]

Warriors <- Warriors[(Warriors$"Games Played" >= 10),]
Warriors_by_starts <- Warriors[order(Warriors$"Games Started",decreasing=TRUE),]
Warriors_starters <- Warriors_by_starts[1:5,]
Warriors_bench <- Warriors_by_starts[6:16,]

Games Played by Starter and Bench Unit Players

Code

ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Games Played")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Games Played", x = "Player", y = "Number of Games Played")

Code

mean(Celtics_starters$"Games Played")

[1] 68.6

Code

ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Games Played")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Games Played", x = "Player", y = "Number of Games Played")

Code

mean(Warriors_starters$"Games Played")

[1] 68.2

Code

ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Games Played")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Games Played", x = "Player", y = "Number of Games Played")

Code

mean(Celtics_bench$"Games Played")

[1] 34.6

Code

ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Games Played")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Games Played", x = "Player", y = "Number of Games Played")

Code

mean(Warriors_bench$"Games Played")

[1] 51.90909

Minutes Played by Starter and Bench Unit Players

Code

ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Minutes Played")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Minutes Played", x = "Player", y = "Number of Minutes Played")

Code

mean(Celtics_starters$"Minutes Played")

[1] 32.1

Code

ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Minutes Played")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Minutes Played", x = "Player", y = "Number of Minutes Played")

Code

mean(Warriors_starters$"Minutes Played")

[1] 29.28

Code

ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Minutes Played")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Minutes Played", x = "Player", y = "Number of Minutes Played")

Code

mean(Celtics_bench$"Minutes Played")

[1] 14.22667

Code

ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Minutes Played")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Minutes Played", x = "Player", y = "Number of Minutes Played")

Code

mean(Warriors_bench$"Minutes Played")

[1] 16.76364

Three Point Shooting Efficiency of Starters and Bench Unit Players

Code

Celtics_starters_no_center <- Celtics_starters[Celtics_starters$Player != 'Robert Williams',]

ggplot(data=Celtics_starters_no_center, aes(x=Celtics_starters_no_center$"Player", y=Celtics_starters_no_center$"Three Point Percentage")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Three Point Percentage", x = "Player", y = "Number of Three Point Percentage")

Code

mean(Celtics_starters_no_center$"Three Point Percentage")

[1] 0.3445

Code

Warriors_starters_no_center <- Warriors_starters[Warriors_starters$Player != 'Kevon Looney',]

ggplot(data=Warriors_starters_no_center, aes(x=Warriors_starters_no_center$"Player", y=Warriors_starters_no_center$"Three Point Percentage")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Three Point Percentage", x = "Player", y = "Number of Three Point Percentage")

Code

mean(Warriors_starters_no_center$"Three Point Percentage")

[1] 0.35825

Code

Celtics_bench_no_center <- Celtics_bench[Celtics_bench$Player != 'Luke Cornet',]

ggplot(data=Celtics_bench_no_center, aes(x=Celtics_bench_no_center$"Player", y=Celtics_bench_no_center$"Three Point Percentage")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Three Point Percentage", x = "Player", y = "Number of Three Point Percentage")

Code

mean(Celtics_bench_no_center$"Three Point Percentage")

[1] 0.3714667

Code

ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Three Point Percentage")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Three Point Percentage", x = "Player", y = "Number of Three Point Percentage")

Code

mean(Warriors_bench$"Three Point Percentage")

[1] 0.3259091

Minutes Played by Starter and Bench Unit Players

Code

ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Two Point Percentage")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Two Point Percentage", x = "Player", y = "Number of Two Point Percentage")

Code

mean(Celtics_starters$"Two Point Percentage")

[1] 0.5788

Code

ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Two Point Percentage")) + geom_bar(stat="identity") +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Two Point Percentage", x = "Player", y = "Number of Two Point Percentage")

Code

mean(Warriors_starters$"Two Point Percentage")

[1] 0.5498

Code

ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Two Point Percentage")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Two Point Percentage", x = "Player", y = "Number of Two Point Percentage")

Code

mean(Celtics_bench$"Two Point Percentage")

[1] 0.5332

Code

ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Two Point Percentage")) + geom_bar(stat="identity") + 
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Two Point Percentage", x = "Player", y = "Number of Two Point Percentage")

Code

mean(Warriors_bench$"Two Point Percentage")

[1] 0.5613636

Critical Reflection

Based on the above analysis, the main difference between the Celtics and the Warriors was the greater contribution of the Warriors bench, who averaged a much larger number of games played per player and greater minutes played. This finding tracks with one of the aspects of the Celtics-Warriors NBA Finals that was discussed, the poor performance of Jayson Tatum, the Celtics’ best player. Because the Celtics’ bench players averaged less games played and minutes played per game, Tatum spent more time on the floor and reported being mentally and physically fatigued during the duration of the series. Clearly Warriors’ bench and their prominent contribution played a major role in their Finals victory during the 2021-2022 season.

My analysis of this dataset does not address the less quantifiable aspects of these basketball teams, such as chemistry between players and the influence of coaches or home-town fans. It would be interesting to explore the relationship between the performance of starters and bench players as it relates to these other variables, such as the difference in scoring efficiency for bench players at their home-court versus during away games. In putting together this project, I learned that additional adjustments to datasets that you didn’t expect are often necessary, such as the removal of players that don’t shoot threes from the analysis of three point efficiency. Overall, it would be fascinating to compare the statistics of the Celtics and Warriors following this season, where the Celtics are off to a fantastic start and the Warriors hover in the middle of their division to see if these findings hold true.

Bibliography

R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.

Wickham, H., & Grolemund, G. (2016). R for data science: Visualize, model, transform, tidy, and import data. OReilly Media.

Etocco. “NBA_Player_Stats.csv.” “Data.world.” 2022. https://data.world/etocco/nba-player-stats

--- title: "Final Project" author: "Daniel Manning" description: "Celtics 2021-2022" date: "2/5/2023" format: html: toc: true code-fold: true code-copy: true code-tools: true categories: - Final_Project --- ```{r} #| label: setup #| warning: false #| message: false library(tidyverse) library(ggplot2) library(here) knitr::opts_chunk$set(echo = TRUE, warning=FALSE, message=FALSE) ``` ## Description of Dataset The NBA Player Stats dataset was sourced from Basketball-Reference.com, a reputable source for broad categories of professional basketball statistics. These statistics are collected by the leagues statisticians through the season and are regarded as very reliable. The dataset includes statistics for every player in the NBA (2374) from the 1997-1998 season to the 2021-2022 season, across 38 teams, so it contains a massive amount of information. This information is broken down into 14573 player-season instances (rows) described by 31 variables (columns). The variable keys and descriptions are listed below for explanation along with their type and cases. Rk: index Player: player name (2374) Pos: position (string) (5 different main cases, 11 combination cases) Tm: team (string) (38) G: games played (numeric) GS: games started (numeric) MP: minutes played per game (numeric) FG: field goals made per game (numeric) FGA: field goals attempted per game (numeric) FG%: field goal percentage (numeric) 3P: three-pointers made per game (numeric) 3PA: three-pointers attempted per game (numeric) 3P%: three-point percentage (numeric) 2P: two-pointers made per game (numeric) 2PA: two-pointers attempted per game (numeric) 2P%: two-point percentage (numeric) eFG%: effective field goal percentage (numeric) FT: free throws made per game (numeric) FTA: free throws attempted per game (numeric) FT%: free throw percentage (numeric) ORB: offensive rebounds per game (numeric) DRB: defensive rebounds per game (numeric) TRB: total rebounds per game (numeric) AST: assists per game (numeric) STL: steals per game (numeric) BLK: blocks per game (numeric) TOV: turnovers per game (numeric) PF: personal fouls per game (numeric) PTS: points per game (numeric) Year: NBA season year (ordinal) (25) ## Load Dataset, Obtain Dimensions, and Unique Cases ```{r} NBA<- here("posts","_data","NBA_Player_Stats.csv")%>% read_csv() dim(NBA) n_distinct(NBA$Player) n_distinct(NBA$Pos) n_distinct(NBA$Tm) n_distinct(NBA$Year) ``` ## Selecting subsets of the data I chose to focus on data from the Celtics' and Warriors' 2021-2022 season in order to compare the two teams who appeared in the NBA finals across some of the major statistics included in the NBA_player_Stats dataset. ```{r} last_season <- NBA[NBA$Year == '2021-2022',] Celtics <- last_season[last_season$Tm == 'BOS',] Warriors <- last_season[last_season$Tm == 'GSW',] ``` ## Change Names of Columns into Recognizable Values/Statistics Next, I recoded the variable keys into the categories/statistics that they describe, for ease of interpretation. ```{r} colnames(Celtics) <- c("Index", "Player", "Position", "Age", "Team", "Games Played", "Games Started", "Minutes Played", "Field Goals Made", "Field Goals Attempted", "Field Goal Percentage", "Three Point Made", "Three Point Attempted", "Three Point Percentage", "Two Point Made", "Two Point Attempted", "Two Point Percentage", "Effective Field Goal Percentage", "Free Throws Made", "Free Throws Attempted", "Free Throw Percentage", "Offensive Rebounds", "Defensive Rebounds", "Total Rebounds", "Assists", "Steals", "Blocks", "Turnovers", "Personal Fouls", "Points", "Season") colnames(Warriors) <- c("Index", "Player", "Position", "Age", "Team", "Games Played", "Games Started", "Minutes Played", "Field Goals Made", "Field Goals Attempted", "Field Goal Percentage", "Three Point Made", "Three Point Attempted", "Three Point Percentage", "Two Point Made", "Two Point Attempted", "Two Point Percentage", "Effective Field Goal Percentage", "Free Throws Made", "Free Throws Attempted", "Free Throw Percentage", "Offensive Rebounds", "Defensive Rebounds", "Total Rebounds", "Assists", "Steals", "Blocks", "Turnovers", "Personal Fouls", "Points", "Season") ``` ## Main Research Question and Subsetting of Data My research question focuses the differences between the Celtics and Warriors starters and bench player statistics. Analysts often focus on the performance of these two groups when discussing team match-ups, as teams often differ in the contribution of their starters and bench, in terms of efficiency, minutes played, etc. Teams with greater contribution from their bench players are said to have greater depth, which can allow a team to withstand injuries throughout the season, for top players to perform optimally, and for coaches to adjust personnel for specific match-ups. In order to avoid players with limited games played to bias the statistics collected, I removed players who played less than ten games. I then ordered the players from each team by descending order of games started, selected the first five as the starting unit, and the remaining players as the bench unit. ```{r} Celtics <- Celtics[(Celtics$"Games Played" >= 10),] Celtics_by_starts <- Celtics[order(Celtics$"Games Started",decreasing=TRUE),] Celtics_starters <- Celtics_by_starts[1:5,] Celtics_bench <- Celtics_by_starts[6:20,] Warriors <- Warriors[(Warriors$"Games Played" >= 10),] Warriors_by_starts <- Warriors[order(Warriors$"Games Started",decreasing=TRUE),] Warriors_starters <- Warriors_by_starts[1:5,] Warriors_bench <- Warriors_by_starts[6:16,] ``` ## Games Played by Starter and Bench Unit Players ```{r} ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Games Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Games Played", x = "Player", y = "Number of Games Played") mean(Celtics_starters$"Games Played") ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Games Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Games Played", x = "Player", y = "Number of Games Played") mean(Warriors_starters$"Games Played") ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Games Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Games Played", x = "Player", y = "Number of Games Played") mean(Celtics_bench$"Games Played") ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Games Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Games Played", x = "Player", y = "Number of Games Played") mean(Warriors_bench$"Games Played") ``` ## Minutes Played by Starter and Bench Unit Players ```{r} ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Minutes Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Minutes Played", x = "Player", y = "Number of Minutes Played") mean(Celtics_starters$"Minutes Played") ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Minutes Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Minutes Played", x = "Player", y = "Number of Minutes Played") mean(Warriors_starters$"Minutes Played") ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Minutes Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Minutes Played", x = "Player", y = "Number of Minutes Played") mean(Celtics_bench$"Minutes Played") ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Minutes Played")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Minutes Played", x = "Player", y = "Number of Minutes Played") mean(Warriors_bench$"Minutes Played") ``` ## Three Point Shooting Efficiency of Starters and Bench Unit Players ```{r} Celtics_starters_no_center <- Celtics_starters[Celtics_starters$Player != 'Robert Williams',] ggplot(data=Celtics_starters_no_center, aes(x=Celtics_starters_no_center$"Player", y=Celtics_starters_no_center$"Three Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Three Point Percentage", x = "Player", y = "Number of Three Point Percentage") mean(Celtics_starters_no_center$"Three Point Percentage") Warriors_starters_no_center <- Warriors_starters[Warriors_starters$Player != 'Kevon Looney',] ggplot(data=Warriors_starters_no_center, aes(x=Warriors_starters_no_center$"Player", y=Warriors_starters_no_center$"Three Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Three Point Percentage", x = "Player", y = "Number of Three Point Percentage") mean(Warriors_starters_no_center$"Three Point Percentage") Celtics_bench_no_center <- Celtics_bench[Celtics_bench$Player != 'Luke Cornet',] ggplot(data=Celtics_bench_no_center, aes(x=Celtics_bench_no_center$"Player", y=Celtics_bench_no_center$"Three Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Three Point Percentage", x = "Player", y = "Number of Three Point Percentage") mean(Celtics_bench_no_center$"Three Point Percentage") ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Three Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Three Point Percentage", x = "Player", y = "Number of Three Point Percentage") mean(Warriors_bench$"Three Point Percentage") ``` ## Minutes Played by Starter and Bench Unit Players ```{r} ggplot(data=Celtics_starters, aes(x=Celtics_starters$"Player", y=Celtics_starters$"Two Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) +labs(title = "Celtics Starters Two Point Percentage", x = "Player", y = "Number of Two Point Percentage") mean(Celtics_starters$"Two Point Percentage") ggplot(data=Warriors_starters, aes(x=Warriors_starters$"Player", y=Warriors_starters$"Two Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1))+ labs(title = "Warriors Starters Two Point Percentage", x = "Player", y = "Number of Two Point Percentage") mean(Warriors_starters$"Two Point Percentage") ggplot(data=Celtics_bench, aes(x=Celtics_bench$"Player", y=Celtics_bench$"Two Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Celtics Bench Two Point Percentage", x = "Player", y = "Number of Two Point Percentage") mean(Celtics_bench$"Two Point Percentage") ggplot(data=Warriors_bench, aes(x=Warriors_bench$"Player", y=Warriors_bench$"Two Point Percentage")) + geom_bar(stat="identity") + theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust=1)) + labs(title = "Warriors Bench Two Point Percentage", x = "Player", y = "Number of Two Point Percentage") mean(Warriors_bench$"Two Point Percentage") ``` ## Critical Reflection Based on the above analysis, the main difference between the Celtics and the Warriors was the greater contribution of the Warriors bench, who averaged a much larger number of games played per player and greater minutes played. This finding tracks with one of the aspects of the Celtics-Warriors NBA Finals that was discussed, the poor performance of Jayson Tatum, the Celtics' best player. Because the Celtics' bench players averaged less games played and minutes played per game, Tatum spent more time on the floor and reported being mentally and physically fatigued during the duration of the series. Clearly Warriors' bench and their prominent contribution played a major role in their Finals victory during the 2021-2022 season. My analysis of this dataset does not address the less quantifiable aspects of these basketball teams, such as chemistry between players and the influence of coaches or home-town fans. It would be interesting to explore the relationship between the performance of starters and bench players as it relates to these other variables, such as the difference in scoring efficiency for bench players at their home-court versus during away games. In putting together this project, I learned that additional adjustments to datasets that you didn't expect are often necessary, such as the removal of players that don't shoot threes from the analysis of three point efficiency. Overall, it would be fascinating to compare the statistics of the Celtics and Warriors following this season, where the Celtics are off to a fantastic start and the Warriors hover in the middle of their division to see if these findings hold true. ## Bibliography R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Wickham, H., & Grolemund, G. (2016). R for data science: Visualize, model, transform, tidy, and import data. OReilly Media. Etocco. "NBA_Player_Stats.csv." "Data.world." 2022. <https://data.world/etocco/nba-player-stats>