Zone Vineyard 8/2

strike_zone_limits <- list(
  xmin = -0.85,
  xmax = 0.85,
  ymin = 1.5,
  ymax = 3.5
)


umpire_data <- keene_data %>%
  filter(PitchCall %in% c("BallCalled", "StrikeCalled")) %>%
  select(Pitcher, Batter, PitchCall, PlateLocHeight, PlateLocSide, PitcherTeam) %>%
  mutate(
    CorrectCall = case_when(
      PitchCall == "BallCalled" & (PlateLocSide < strike_zone_limits$xmin | PlateLocSide > strike_zone_limits$xmax | PlateLocHeight < strike_zone_limits$ymin | PlateLocHeight > strike_zone_limits$ymax) ~ "Correct",
      PitchCall == "StrikeCalled" & (PlateLocSide >= strike_zone_limits$xmin & PlateLocSide <= strike_zone_limits$xmax & PlateLocHeight >= strike_zone_limits$ymin & PlateLocHeight <= strike_zone_limits$ymax) ~ "Correct",
      TRUE ~ "Incorrect"
    ),
    Distance = case_when(
      PitchCall == "BallCalled" & PlateLocSide < strike_zone_limits$xmin & PlateLocHeight >= strike_zone_limits$ymin & PlateLocHeight <= strike_zone_limits$ymax ~ (strike_zone_limits$xmin - PlateLocSide) * 12,
      PitchCall == "BallCalled" & PlateLocSide > strike_zone_limits$xmax & PlateLocHeight >= strike_zone_limits$ymin & PlateLocHeight <= strike_zone_limits$ymax ~ (PlateLocSide - strike_zone_limits$xmax) * 12,
      PitchCall == "BallCalled" & PlateLocHeight < strike_zone_limits$ymin & PlateLocSide >= strike_zone_limits$xmin & PlateLocSide <= strike_zone_limits$xmax ~ (strike_zone_limits$ymin - PlateLocHeight) * 12,
      PitchCall == "BallCalled" & PlateLocHeight > strike_zone_limits$ymax & PlateLocSide >= strike_zone_limits$xmin & PlateLocSide <= strike_zone_limits$xmax ~ (PlateLocHeight - strike_zone_limits$ymax) * 12,
      PitchCall == "BallCalled" & (PlateLocSide < strike_zone_limits$xmin | PlateLocSide > strike_zone_limits$xmax) & (PlateLocHeight < strike_zone_limits$ymin | PlateLocHeight > strike_zone_limits$ymax) ~ sqrt((pmax(0, strike_zone_limits$xmin - PlateLocSide, PlateLocSide - strike_zone_limits$xmax))^2 + (pmax(0, strike_zone_limits$ymin - PlateLocHeight, PlateLocHeight - strike_zone_limits$ymax))^2) * 12,
      
      # Strikes called as balls
      PitchCall == "StrikeCalled" ~ sqrt((pmax(0, strike_zone_limits$xmin - PlateLocSide, PlateLocSide - strike_zone_limits$xmax))^2 + (pmax(0, strike_zone_limits$ymin - PlateLocHeight, PlateLocHeight - strike_zone_limits$ymax))^2) * 12
    )
  )

plot <- ggplot(umpire_data, aes(x = PlateLocSide, y = PlateLocHeight, color = PitchCall, text = paste("Pitcher: ", Pitcher, "<br>Batter: ", Batter, "<br>Distance: ", round(Distance, 2), " inches"))) +
  geom_point(data = umpire_data %>% filter(CorrectCall == "Correct"), size = 1, shape = 16) + # Correct calls as smaller dots
  geom_point(data = umpire_data %>% filter(CorrectCall == "Incorrect"), size = 2, shape = 4, stroke = 2) + 
  geom_rect(aes(xmin = strike_zone_limits$xmin, xmax = strike_zone_limits$xmax, ymin = strike_zone_limits$ymin, ymax = strike_zone_limits$ymax), fill = NA, color = "black", linetype = "solid", size = 1) + # Strike zone 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 = "Umpire's Ball and Strike Calls",
       x = "Horizontal Location (feet)",
       y = "Vertical Location (feet)",
       color = "Pitch Call") +
  scale_color_manual(values = c("BallCalled" = "green", "StrikeCalled" = "red")) +
  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"),
    plot.margin = unit(c(2, 2, 2, 2), "cm") # Increase plot margins
  )

plotly_plot <- ggplotly(plot, tooltip = "text") %>%
  layout(clickmode = 'event+select') # Require clicking to see the data

plotly_plot

# Load necessary libraries
library(readxl)
library(dplyr)
library(kableExtra)

# Load the data
file_path <- "C:\\Users\\Franco Castagliuolo\\OneDrive - Bentley University\\Vineyard82.xlsx"
keene_data <- read_excel(file_path)

# Define strike zone limits
strike_zone_limits <- list(
  xmin = -0.85,
  xmax = 0.85,
  ymin = 1.5,
  ymax = 3.5
)

# Filter data and calculate correct percentages for Vermont and Opponent
vermont_summary <- keene_data %>%
  filter(PitchCall %in% c("BallCalled", "StrikeCalled"), PitcherTeam == "VER_MOU") %>%
  filter(!is.na(PlateLocSide) & !is.na(PlateLocHeight)) %>%
  mutate(
    TrueBall = PlateLocSide < strike_zone_limits$xmin | PlateLocSide > strike_zone_limits$xmax | PlateLocHeight < strike_zone_limits$ymin | PlateLocHeight > strike_zone_limits$ymax,
    TrueStrike = PlateLocSide >= strike_zone_limits$xmin & PlateLocSide <= strike_zone_limits$xmax & PlateLocHeight >= strike_zone_limits$ymin & PlateLocHeight <= strike_zone_limits$ymax
  ) %>%
  summarise(
    Team = "VER_MOU",
    Total_TrueBalls = sum(TrueBall),
    Correct_BallCalls = sum(PitchCall == "BallCalled" & TrueBall),
    Total_TrueStrikes = sum(TrueStrike),
    Correct_StrikeCalls = sum(PitchCall == "StrikeCalled" & TrueStrike),
    Correct_Percentage_BallCalled = ifelse(Total_TrueBalls > 0, Correct_BallCalls / Total_TrueBalls * 100, NA),
    Correct_Percentage_StrikeCalled = ifelse(Total_TrueStrikes > 0, Correct_StrikeCalls / Total_TrueStrikes * 100, NA)
  )

opponent_summary <- keene_data %>%
  filter(PitchCall %in% c("BallCalled", "StrikeCalled"), PitcherTeam != "VER_MOU") %>%
  filter(!is.na(PlateLocSide) & !is.na(PlateLocHeight)) %>%
  mutate(
    TrueBall = PlateLocSide < strike_zone_limits$xmin | PlateLocSide > strike_zone_limits$xmax | PlateLocHeight < strike_zone_limits$ymin | PlateLocHeight > strike_zone_limits$ymax,
    TrueStrike = PlateLocSide >= strike_zone_limits$xmin & PlateLocSide <= strike_zone_limits$xmax & PlateLocHeight >= strike_zone_limits$ymin & PlateLocHeight <= strike_zone_limits$ymax
  ) %>%
  summarise(
    Team = "Opponent",
    Total_TrueBalls = sum(TrueBall),
    Correct_BallCalls = sum(PitchCall == "BallCalled" & TrueBall),
    Total_TrueStrikes = sum(TrueStrike),
    Correct_StrikeCalls = sum(PitchCall == "StrikeCalled" & TrueStrike),
    Correct_Percentage_BallCalled = ifelse(Total_TrueBalls > 0, Correct_BallCalls / Total_TrueBalls * 100, NA),
    Correct_Percentage_StrikeCalled = ifelse(Total_TrueStrikes > 0, Correct_StrikeCalls / Total_TrueStrikes * 100, NA)
  )

# Combine the summaries
combined_summary <- bind_rows(vermont_summary, opponent_summary)

# Display the summary table
kable(combined_summary, caption = "Summary of Correct Ball and Strike Calls") %>%
  kable_styling(full_width = F, position = "left")

Summary of Correct Ball and Strike Calls

Team	Total_TrueBalls	Correct_BallCalls	Total_TrueStrikes	Correct_StrikeCalls	Correct_Percentage_BallCalled	Correct_Percentage_StrikeCalled
VER_MOU	71	65	26	22	91.54930	84.61538
Opponent	88	84	41	38	95.45455	92.68293

Correct_Percentage_BallCalled: - Percentage of balls called correctly from the pitcher’s perspective - The Pitching team would prefer a lower percentage - Correct is a pitch outside the zone called a ball - Incorrect is a pitch outside the zone called as a strike

Correct_Percentage_StrikeCalled - Percentage of strikes called correctly from the pitcher’s perspective - The pitching team would prefer a higher percentage - Correct is a pitch inside the zone called as a strike - Incorrect is a pitch inside the zone called as a ball

The pitching team would prefer:

A lower percentage of accurate ball calls because this indicates more balls are incorrectly called as strikes.

A higher percentage of accurate strike calls because this indicates fewer strikes are incorrectly called as balls.