MLB Batted Ball Data Analysis
Aiden Scholey
2023-07-18
This is an R Markdown document analyzing Major League Baseball (MLB) batting data from 2015-2022
This data is uploaded on Kaggle and contains several variables that pertain to the quality of batter contact.
Some of these variables are self-explanatory such as homerun distance and exit velocity. Greater values in these variables are better for batters. However, some of these measures are less self-explanatory for someone who is not a baseball fan. To help, two images from MLB.com have been inserted to help explain what launch angle and sweet spot are, and how to identify what a barrelled ball is.
MLB.com explanation of sweet spot and launch
angle.
From MLB.com “A player who hits the ball solidly is said to have gotten the”sweet spot” of the bat on the ball. The sweet spot classification quantifies that as a batted-ball event with a launch angle ranging from 8 to 32 degrees.”
MLB.com illustration of the exit velocity and
launch angle combinations that yield Barreled status in 100 percent of
cases.
From MLB.com “To be Barreled, a batted ball requires an exit velocity of at least 98 mph. At that speed, balls struck with a launch angle between 26-30 degrees always garner Barreled classification. For every mph over 98, the range of launch angles expands.”
Installing and Loading Packages
For this project we should use these packages to to clean and visualize our dataset.
pacman::p_load(tidyverse,rio,ggplot2,ggpubr,ggsci,vtable,rmarkdown)Loading Data
Next we can use the read_csv function to load our dataset into R. To save extra typing, the dataset will be referred to as “mlb” from now on and is saved as a variable under that name.
mlb = read.csv('https://www.kaggle.com/datasets/mattop/mlb-batting-exit-velocity-data-2015-2022?select=mlb-batter-exit-velocity.csv')
mlb_batter_exit_velocity <- read_csv("R_Files/R Course Files/mlb-batter-exit-velocity.csv")## Rows: 1983 Columns: 20
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): player
## dbl (19): id, rank, year, batted_ball_events, launch_angle, sweet_spot_perce...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.mlb = mlb_batter_exit_velocityView Data
We can run some summary statistics on the MLB data. The first function gives some descriptive statistics for the data and the second function creates a nice table for us to view the descriptive statistics in. Some of the variables such as ID, Rank and Year are less useful/interesting to get summary statistics on and can be excluded from this section by creating a second dataframe that removes these columns. We will only use that dataframe for this table.
mlb2 = mlb[,-1:-3]
summary(mlb2)## player batted_ball_events launch_angle sweet_spot_percentage
## Length:1983 Min. : 67.0 Min. :-7.90 Min. :18.50
## Class :character 1st Qu.:256.0 1st Qu.: 9.50 1st Qu.:30.90
## Mode :character Median :333.0 Median :12.50 Median :33.50
## Mean :328.7 Mean :12.53 Mean :33.59
## 3rd Qu.:415.0 3rd Qu.:15.55 3rd Qu.:36.30
## Max. :599.0 Max. :26.80 Max. :49.70
##
## max_ev average_ev fly_ball_line_drive_ev ground_ball_ev
## Min. : 98.5 Min. :78.20 Min. : 83.00 Min. :71.20
## 1st Qu.:108.6 1st Qu.:87.35 1st Qu.: 91.20 1st Qu.:84.10
## Median :110.6 Median :88.80 Median : 92.90 Median :85.90
## Mean :110.8 Mean :88.77 Mean : 92.81 Mean :85.86
## 3rd Qu.:112.9 3rd Qu.:90.10 3rd Qu.: 94.50 3rd Qu.:87.60
## Max. :122.2 Max. :96.00 Max. :101.30 Max. :94.10
##
## max_distance average_distance average_homerun hard_hit_95mph+
## Min. :359 Min. : 98.0 Min. :342.0 Min. : 6.0
## 1st Qu.:423 1st Qu.:158.0 1st Qu.:392.0 1st Qu.: 87.5
## Median :436 Median :172.0 Median :399.0 Median :120.0
## Mean :436 Mean :171.4 Mean :398.3 Mean :122.3
## 3rd Qu.:449 3rd Qu.:186.0 3rd Qu.:406.0 3rd Qu.:157.0
## Max. :505 Max. :234.0 Max. :438.0 Max. :274.0
## NA's :9
## hard_hit_percentage hard_hit_swing_percentage total_barrels
## Min. : 6.40 Min. : 2.30 Min. : 0.00
## 1st Qu.:32.80 1st Qu.:12.30 1st Qu.:12.00
## Median :38.20 Median :14.20 Median :21.00
## Mean :37.57 Mean :14.14 Mean :23.92
## 3rd Qu.:42.70 3rd Qu.:16.00 3rd Qu.:33.00
## Max. :62.20 Max. :23.90 Max. :91.00
##
## barrels_batted_balls_percentage barrels_plate_appearance_percentage
## Min. : 0.000 Min. : 0.000
## 1st Qu.: 4.500 1st Qu.: 3.200
## Median : 7.100 Median : 4.900
## Mean : 7.521 Mean : 5.021
## 3rd Qu.:10.150 3rd Qu.: 6.700
## Max. :26.700 Max. :15.800
## st(mlb2,vars = , title = "MLB Exit Velocity Summary Statistics")| Variable | N | Mean | Std. Dev. | Min | Pctl. 25 | Pctl. 75 | Max |
|---|---|---|---|---|---|---|---|
| batted_ball_events | 1983 | 329 | 113 | 67 | 256 | 415 | 599 |
| launch_angle | 1983 | 13 | 4.6 | -7.9 | 9.5 | 16 | 27 |
| sweet_spot_percentage | 1983 | 34 | 4.1 | 18 | 31 | 36 | 50 |
| max_ev | 1983 | 111 | 3.2 | 98 | 109 | 113 | 122 |
| average_ev | 1983 | 89 | 2.2 | 78 | 87 | 90 | 96 |
| fly_ball_line_drive_ev | 1983 | 93 | 2.6 | 83 | 91 | 94 | 101 |
| ground_ball_ev | 1983 | 86 | 2.6 | 71 | 84 | 88 | 94 |
| max_distance | 1983 | 436 | 20 | 359 | 423 | 449 | 505 |
| average_distance | 1983 | 171 | 20 | 98 | 158 | 186 | 234 |
| average_homerun | 1974 | 398 | 11 | 342 | 392 | 406 | 438 |
| hard_hit_95mph+ | 1983 | 122 | 48 | 6 | 88 | 157 | 274 |
| hard_hit_percentage | 1983 | 38 | 7.9 | 6.4 | 33 | 43 | 62 |
| hard_hit_swing_percentage | 1983 | 14 | 2.9 | 2.3 | 12 | 16 | 24 |
| total_barrels | 1983 | 24 | 15 | 0 | 12 | 33 | 91 |
| barrels_batted_balls_percentage | 1983 | 7.5 | 4.2 | 0 | 4.5 | 10 | 27 |
| barrels_plate_appearance_percentage | 1983 | 5 | 2.5 | 0 | 3.2 | 6.7 | 16 |
Cleaning Data with dplyr
#This line of code renames a few variables, to simplify a few of their names.
mlb = rename(mlb, hard_hit_count = "hard_hit_95mph+", average_homerun_distance = "average_homerun", average_hit_distance = "average_distance")Histogram & Density Plots
#Plot for average exit velocity
ggplot(data = mlb, aes(x = average_ev)) +
geom_histogram(bins = 20, aes(y =..density..),color = "black",fill = "white")+
geom_density(alpha = .2, fill="#ff6666")+
scale_y_continuous(name = "Density",sec.axis = sec_axis(~.*1500, name = "Count", breaks = seq(0,350,50)))+
theme(axis.title.y.left = element_text(vjust = +3),axis.title.y.right = element_text(vjust = +3))+
xlab("Average Exit Velocity (mph)")+
ylab("Density")+
ggtitle("Average Exit Velocity Distribution")## Warning: The dot-dot notation (`..density..`) was deprecated in ggplot2 3.4.0.
## ℹ Please use `after_stat(density)` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
#Plot for max homerun distance
ggplot(data = mlb, aes(x = max_distance)) +
geom_histogram(bins = 25, aes(y =..density..),color = "darkblue",fill = "white")+
geom_density(alpha = .2, fill="cyan")+
scale_y_continuous(name = "Density",sec.axis = sec_axis(~.*25000, name = "Count", breaks = seq(0,500,100)))+
theme(axis.title.y.left = element_text(vjust = +3),axis.title.y.right = element_text(vjust = +3))+
xlab("Maximum Homerun Distance (ft)")+
ylab("Density")+
ggtitle("Maximum Homerun Distance Distribution")
#Plot for launch angle
ggplot(data = mlb, aes(x = launch_angle)) +
geom_histogram(bins = 20, aes(y = ..density..), color = "purple", fill = "white")+
geom_density(alpha = .2, fill = "magenta")+
scale_y_continuous(name = "Density",sec.axis = sec_axis(~.*3500, name = "Count", breaks = seq(0,350,50)))+
theme(axis.title.y.left = element_text(vjust = +3),axis.title.y.right = element_text(vjust = +3))+
xlab("Launch Angle (°)")+
ylab("Density")+
ggtitle("Launch Angle Distribution")
Histogram + Boxplot Graphs
plot1 = ggplot(data = mlb, aes(x = hard_hit_percentage))+
geom_boxplot(fill = "lightblue", color = "black")+
xlab("Hard Hit Percentage")+
theme_classic()+
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank()
)
plot2 = ggplot(data = mlb, aes(x = hard_hit_percentage))+
geom_histogram(bins = 30, aes(x = hard_hit_percentage), fill = "lightblue",color = "black")+
ylab("Relative Frequency")+
xlab("Hard Hit Percentage")+
theme_classic()
cowplot::plot_grid(plot2,plot1,
ncol = 1, rel_heights = c(2,1),
align = "v",axis = "lr")
plot3 = ggplot(data = mlb, aes(x = sweet_spot_percentage))+
geom_boxplot(fill = "gold", color = "black")+
xlab("Sweet Spot Percentage")+
theme(axis.ticks.y = element_blank(),axis.text.y = element_blank())+
theme_classic()+
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank()
)
plot4 = ggplot(data = mlb, aes(x = sweet_spot_percentage))+
geom_histogram(bins = 30, aes(x = hard_hit_percentage), fill = "gold", color = "black")+
xlab("Sweet Spot Percentage")+
ylab("Relative Frequency")+
theme_classic()
cowplot::plot_grid(plot4,plot3,
ncol = 1, rel_heights = c(4,3),
align = "v",axis = "lr")
Scatter Plots
ggplot(data = mlb,mapping = aes(x = hard_hit_percentage,y=average_ev))+
geom_point(aes(color=max_ev))+
geom_smooth(method = lm)+
stat_regline_equation(label.y = 98, aes(label = ..eq.label..))+
stat_regline_equation(label.y = 96, aes(label = ..rr.label..))+
scale_color_gradient(low = "blue",high = "red")+
guides(fill = guide_legend(
direction = "horizontal",
title.position = "top",
label.position = "bottom",
label.hjust = 1))+
labs(title = "Hard Hit Percentage vs. Average Exit Velocity")+
labs(x= "Hard Hit Percentage (%)",y ="Average Exit Velocity (mph)",color = "Maximum Exit Velocity")## `geom_smooth()` using formula = 'y ~ x'
ggplot(data = mlb, mapping = aes(x = hard_hit_count, y = total_barrels))+
geom_point(aes(color=max_ev))+
geom_smooth(method = lm)+
theme_classic()+
stat_regline_equation(label.y = 89, aes(label = ..eq.label..))+
stat_regline_equation(label.y = 82, aes(label = ..rr.label..))+
scale_color_gradient(low = "blue",high = "red")+
guides(fill = guide_legend(
direction = "horizontal",
title.position = "top",
label.position = "bottom",
label.hjust = 1))+
labs(title = "Hard Hit Count vs. Total Barrels")+
labs(x = "Hard Hit Count", y = "Total Barrels", color = "Maximum Exit Velocity")## `geom_smooth()` using formula = 'y ~ x'