MLB EV vs SLG 2023

install.packages("baseballr")
install.packages("mlbplotR")
library(tidyverse)
library(dplyr)
library(readr)
library(ggplot2)
library(ggrepel)
library(tidyr)
library(ggimage)
library(hockeyR)
library(mlbplotR)
library(ggimage)
help(package = "baseballr")
tictoc::tic()
progressr::with_progress({
  mlb_stat_names <- baseballr::mlb_baseball_stats() #select season
})
tictoc::toc()

tictoc::tic()
progressr::with_progress({
  mlb_logos <- baseballr::mlb_teams() #select season
})
tictoc::toc()

mlb_logos <- mlb_logos %>%
  filter(league_id %in% c("103", "104"))


mlb_logos <- mlbplotR::load_mlb_teams() |> 
  dplyr::filter(!team_abbr %in% c("AL", "NL", "MLB")) |> 


# Get team batting leaderboard data
tictoc::tic()
progressr::with_progress({
  team_batting_data <- baseballr::fg_team_batter
})
tictoc::toc()

head(team_batting_data())

# Define the date range for the season
start_date <- "2023-03-30"
end_date <- "2023-10-01"

standard_batting <- read.csv("C:\\Users\\imksy\\Documents\\baseball_reference_slg.csv")

advanced_batting <- read.csv("C:\\Users\\imksy\\Documents\\baseball_reference_ev.csv")

# Merge the datasets based on a common identifier (e.g., team)
combined_batting <- merge(standard_batting, advanced_batting, by = c("Tm"))

# Merge the datasets based on a common identifier (e.g., team)
combined_batting_logos <- merge(combined_batting, mlb_logos, by.x = "Tm", by.y = "team_name")

avg_ev <- mean(combined_batting_logos$EV, na.rm = TRUE)
avg_slg <- mean(combined_batting_logos$SLG, na.rm = TRUE)

# Calculate non_HR_slugging
combined_batting_logos$non_HR_slugging <- ((combined_batting_logos$H - (combined_batting_logos$X2B + combined_batting_logos$X3B) + (2 * combined_batting_logos$X2B) + (3 * combined_batting_logos$X3B)) / (combined_batting_logos$AB - combined_batting_logos$HR))
#might need to also adjust for not accounting for ev hits without HR

avg_nonHr_slg <- mean(combined_batting_logos$non_HR_slugging, na.rm = TRUE)

correlation_coefficient_ev_slg <- cor(combined_batting_logos$EV, combined_batting_logos$SLG, use = "complete.obs")
correlation_coefficient_ev_slg_noHR <- cor(combined_batting_logos$EV, combined_batting_logos$non_HR_slugging, use = "complete.obs")

ggplot(combined_batting_logos, aes(x = EV, y = SLG)) +
  geom_image(aes(image = team_logo_espn), size = 0.1) + 
  geom_vline(xintercept = avg_ev, linetype = "dashed") +
  geom_hline(yintercept = avg_slg, linetype = "dashed") +
  labs(title = "Slugging Percentage vs Exit Velocity by Team, 2023 MLB Season",
       x = "Exit Velocity (EV)",
       y = "Slugging Percentage (SLG)",
       color = "Team") +
  scale_x_continuous(
    limits = c(87, 91) #breaks
  ) +
  scale_y_continuous(limits = c(0.37, 0.501)) + 
  theme_minimal() +
    theme(
      plot.title = element_text(hjust = 0.5),
      panel.grid.major = element_line(color = "gray", linetype = "dashed")
    ) +
  annotate("text", x = 91, y = 0.4, label = paste("Correlation:", round(correlation_coefficient_ev_slg, 5)), 
           hjust = 1, vjust = 0, size = 3, color = "red")

ggplot(combined_batting_logos, aes(x = EV, y = non_HR_slugging)) +
  geom_image(aes(image = team_logo_espn), size = 0.1) + 
  geom_vline(xintercept = avg_ev, linetype = "dashed") +
  geom_hline(yintercept = avg_nonHr_slg, linetype = "dashed") +
  labs(title = "Non-HR Slugging Percentage vs Exit Velocity by Team, 2023 MLB Season",
       x = "Exit Velocity (EV)",
       y = "Non-HR Slugging Percentage",
       color = "Team") + 
  theme_minimal() +
  theme(
    plot.title = element_text(hjust = 0.5),
    panel.grid.major = element_line(color = "gray", linetype = "dashed")
  ) +
  annotate("text", x = 91, y = 0.3, label = paste("Correlation:", round(correlation_coefficient_ev_slg_noHR, 5)), 
           hjust = 1, vjust = 0, size = 3, color = "red")

#https://twitter.com/TotallyREALSpo1/status/1794970451715756266