knitr::opts_chunk$set(echo = TRUE, message = FALSE, warning = FALSE)
library(readxl)
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(ggplot2)
library(knitr)
# Load the data
Upper626 <- read_excel("C:/Users/Franco Castagliuolo/OneDrive - Bentley University/Neers 24/Pitchers/Upper 626/Upper 626.xlsx")
# Filter the data for the pitcher Josh Keevan
Josh_Keevan_data <- Upper626 %>%
filter(Pitcher == "Keevan, Josh")
# Create a detailed table for each pitch
detailed_pitch_table <- Josh_Keevan_data %>%
select(AutoPitchType, RelSpeed, SpinRate, SpinAxis, HorzBreak, InducedVertBreak, PitchCall, RelHeight, RelSide, Extension) %>%
rename(
ReleaseSpeed = RelSpeed,
Tilt = SpinAxis,
HorizontalBreak = HorzBreak,
InducedVerticalBreak = InducedVertBreak,
ReleaseHeight = RelHeight,
ReleaseSide = RelSide
) %>%
mutate(
ReleaseSpeed = round(ReleaseSpeed, 2),
SpinRate = round(SpinRate, 2),
Tilt = round(Tilt, 2),
HorizontalBreak = round(HorizontalBreak, 2),
InducedVerticalBreak = round(InducedVerticalBreak, 2),
ReleaseHeight = round(ReleaseHeight, 2),
ReleaseSide = round(ReleaseSide, 2),
Extension = round(Extension, 2),
ClockTilt = round((Tilt / 30) %% 12, 1) # Interpret Tilt as clock face
)
# Display the detailed table
knitr::kable(detailed_pitch_table, caption = "Detailed Pitch Table for Josh Keevan")
Detailed Pitch Table for Josh Keevan
| Changeup |
75.95 |
1716.08 |
142.80 |
-9.12 |
13.54 |
BallCalled |
5.28 |
-0.52 |
6.53 |
4.8 |
| Changeup |
81.91 |
2189.44 |
131.98 |
-17.20 |
16.87 |
StrikeCalled |
5.50 |
-0.49 |
5.55 |
4.4 |
| Sinker |
85.30 |
2235.95 |
137.66 |
-14.35 |
17.02 |
FoulBallNotFieldable |
5.63 |
-0.19 |
5.85 |
4.6 |
| Changeup |
75.19 |
1779.18 |
117.16 |
-14.73 |
9.04 |
InPlay |
5.17 |
-0.62 |
6.06 |
3.9 |
| Changeup |
83.46 |
2193.49 |
128.57 |
-18.64 |
16.25 |
BallCalled |
5.63 |
-0.26 |
5.59 |
4.3 |
| Changeup |
84.62 |
2258.51 |
133.83 |
-15.95 |
16.62 |
BallCalled |
5.54 |
-0.21 |
5.84 |
4.5 |
| Changeup |
82.71 |
2190.38 |
136.26 |
-15.63 |
17.73 |
StrikeSwinging |
5.39 |
-0.49 |
5.80 |
4.5 |
| Splitter |
84.22 |
2173.39 |
139.82 |
-14.49 |
18.53 |
FoulBallNotFieldable |
5.58 |
-0.25 |
5.71 |
4.7 |
| Changeup |
74.65 |
1927.05 |
120.45 |
-14.29 |
10.05 |
StrikeSwinging |
5.20 |
-0.76 |
5.68 |
4.0 |
| Four-Seam |
84.02 |
2280.46 |
141.35 |
-13.14 |
17.90 |
BallCalled |
5.60 |
-0.33 |
5.54 |
4.7 |
| Changeup |
83.05 |
2291.25 |
129.99 |
-16.62 |
15.40 |
InPlay |
5.49 |
-0.36 |
5.76 |
4.3 |
# Calculate the total number of pitches
total_pitches <- nrow(detailed_pitch_table)
# Create a summary table
pitch_summary <- detailed_pitch_table %>%
group_by(AutoPitchType) %>%
summarise(
TotalPitches = n(),
Usage = sprintf("%.2f%%", n() / total_pitches * 100),
Balls = sum(PitchCall == "BallCalled"),
Strikes = sum(PitchCall != "BallCalled"), # Count everything not a ball as a strike
BallPercentage = sprintf('%.2f%%', Balls / TotalPitches * 100),
StrikePercentage = sprintf('%.2f%%', Strikes / TotalPitches * 100),
AvgVelocity = round(mean(ReleaseSpeed, na.rm = TRUE), 2),
AvgSpinRate = round(mean(SpinRate, na.rm = TRUE), 2),
AvgInducedVertBreak = round(mean(InducedVerticalBreak, na.rm = TRUE), 2),
AvgHorzBreak = round(mean(HorizontalBreak, na.rm = TRUE), 2),
AvgTilt = round(mean(Tilt, na.rm = TRUE), 2),
AvgClockTilt = round(mean(ClockTilt, na.rm = TRUE), 1), # Clock face interpretation
AvgReleaseHeight = round(mean(ReleaseHeight, na.rm = TRUE), 2),
AvgReleaseSide = round(mean(ReleaseSide, na.rm = TRUE), 2),
AvgExtension = round(mean(Extension, na.rm = TRUE), 2)
) %>%
select(AutoPitchType, Usage, everything())
# Display the total number of pitches
cat("Total number of pitches thrown: ", total_pitches, "\n")
## Total number of pitches thrown: 11
# Display the summary table
knitr::kable(pitch_summary, caption = "Summary Pitch Table for Josh Keevan")
Summary Pitch Table for Josh Keevan
| Changeup |
72.73% |
8 |
3 |
5 |
37.50% |
62.50% |
80.19 |
2068.17 |
14.44 |
-15.27 |
130.13 |
4.3 |
5.40 |
-0.46 |
5.85 |
| Four-Seam |
9.09% |
1 |
1 |
0 |
100.00% |
0.00% |
84.02 |
2280.46 |
17.90 |
-13.14 |
141.35 |
4.7 |
5.60 |
-0.33 |
5.54 |
| Sinker |
9.09% |
1 |
0 |
1 |
0.00% |
100.00% |
85.30 |
2235.95 |
17.02 |
-14.35 |
137.66 |
4.6 |
5.63 |
-0.19 |
5.85 |
| Splitter |
9.09% |
1 |
0 |
1 |
0.00% |
100.00% |
84.22 |
2173.39 |
18.53 |
-14.49 |
139.82 |
4.7 |
5.58 |
-0.25 |
5.71 |
# Calculate maximum fastball velocity
max_fb_velocity <- detailed_pitch_table %>%
filter(AutoPitchType %in% c("Four-Seam", "Two-Seam", "Sinker", "Cutter")) %>%
summarise(MaxFBVelocity = max(ReleaseSpeed, na.rm = TRUE)) %>%
pull(MaxFBVelocity)
# Display the maximum fastball velocity
cat("Josh Keevan maximum FB velocity: ", max_fb_velocity, "mph\n")
## Josh Keevan maximum FB velocity: 85.3 mph
# Prepare data for plotting pitch locations
pitch_location_data <- Josh_Keevan_data %>%
select(AutoPitchType, PlateLocHeight, PlateLocSide, PitchCall) %>%
rename(
PitchHeight = PlateLocHeight,
PitchSide = PlateLocSide
) %>%
mutate(
SwingTake = ifelse(PitchCall %in% c("StrikeSwinging", "FoulBallNonSwinging", "FoulBallFieldable", "FoulBallNotFieldable", "InPlay"), "Swing", "Take"),
Chase = ifelse(SwingTake == "Swing" & (PitchSide < -0.75 | PitchSide > 0.75 | PitchHeight < 1.5 | PitchHeight > 3.5), "Chase", "Non-Chase")
)
# Create the scatter plot with specified strike zone boxes
ggplot(pitch_location_data, aes(x = PitchSide, y = PitchHeight, color = SwingTake, shape = Chase)) +
geom_point(size = 3) + # Increase point size
geom_rect(aes(xmin = -0.5, xmax = 0.5, ymin = 1.75, ymax = 3.25), fill = NA, color = "red", linetype = "solid", size = 1) + # Red box
geom_rect(aes(xmin = -0.75, xmax = 0.75, ymin = 1.5, ymax = 3.5), fill = NA, color = "black", linetype = "solid", size = 1) + # Black box
geom_rect(aes(xmin = -1.25, xmax = 1.25, ymin = 1.25, ymax = 3.75), fill = NA, color = "gray", linetype = "solid", size = 1) + # Gray box
scale_x_continuous(limits = c(-2, 2)) +
scale_y_continuous(limits = c(0, 5)) +
coord_fixed(ratio = 1) + # Adjust ratio to shrink vertical distance
labs(title = "Pitch Locations for Josh Keevan",
x = "Horizontal Location (feet)",
y = "Vertical Location (feet)",
color = "Swing/Take",
shape = "Chase") +
facet_wrap(~ AutoPitchType) + # Create individual graphs for each pitch type
theme_minimal() +
theme(
legend.position = "right",
panel.grid.major = element_line(color = "grey80"),
panel.grid.minor = element_line(color = "grey90"),
axis.text = element_text(color = "black"),
axis.title = element_text(color = "black"),
plot.title = element_text(color = "black"),
legend.background = element_rect(fill = "white", color = NA),
legend.key = element_rect(fill = "white", color = NA),
legend.text = element_text(color = "black"),
legend.title = element_text(color = "black")
)
# Create the scatter plot for horizontal and vertical breaks
ggplot(detailed_pitch_table, aes(x = HorizontalBreak, y = InducedVerticalBreak, color = AutoPitchType)) +
geom_point(size = 3) + # Increase point size
labs(title = "Pitch Movement for Josh Keevan",
x = "Horizontal Break (inches)",
y = "Induced Vertical Break (inches)",
color = "Pitch Type") +
theme_minimal() +
theme(
legend.position = "right",
panel.grid.major = element_line(color = "grey80"),
panel.grid.minor = element_line(color = "grey90"),
axis.text = element_text(color = "black"),
axis.title = element_text(color = "black"),
plot.title = element_text(color = "black"),
legend.background = element_rect(fill = "white", color = NA),
legend.key = element_rect(fill = "white", color = NA),
legend.text = element_text(color = "black"),
legend.title = element_text(color = "black")
)
