Update app.R

app.R

@@ -1,58 +1,289 @@
+# Load required libraries
 library(shiny)
-library(bslib)
+library(DT)
+library(baseballr)
 library(dplyr)
-library(ggplot2)
-
-df <- readr::read_csv("penguins.csv")
-# Find subset of columns that are suitable for scatter plot
-df_num <- df |> select(where(is.numeric), -Year)
-
-ui <- page_sidebar(
-  theme = bs_theme(bootswatch = "minty"),
-  title = "Penguins explorer",
-  sidebar = sidebar(
-    varSelectInput("xvar", "X variable", df_num, selected = "Bill Length (mm)"),
-    varSelectInput("yvar", "Y variable", df_num, selected = "Bill Depth (mm)"),
-    checkboxGroupInput("species", "Filter by species",
-      choices = unique(df$Species), selected = unique(df$Species)
+library(xgboost)
+
+Sys.setenv(TZ='EST')
+download_private_csv <- function(repo_id, filename) {
+  url <- paste0("https://huggingface.co/datasets/", repo_id, "/resolve/main/", filename)
+  response <- GET(url, add_headers(Authorization = paste("Bearer", Sys.getenv("GETCSV"))))
+  
+  if (status_code(response) == 200) {
+    content <- content(response, "text")
+    con <- textConnection(content)
+    
+    # Try different read options
+    data <- read.csv(con, 
+                     header = TRUE,
+                     check.names = FALSE,  # This prevents R from modifying column names
+                     fileEncoding = "UTF-8",
+                     stringsAsFactors = FALSE)
+    close(con)
+    return(data)
+  } else {
+    stop("Failed to download dataset")
+  }
+}
+MLB <- download_private_csv("TimStats/StatcastDataAll", "MLB.csv")
+AAA <- download_private_csv("TimStats/StatcastDataAll", "AAA.csv")
+FSL <- download_private_csv("TimStats/StatcastDataAll", "FSL.csv")
+
+
+MLB <- MLB %>% 
+  select(
+    `Pitcher Name`, `Pitcher ID`, pitch_name, start_speed, spin_rate, extension,
+    IVB, HB, x0, z0
+  )
+
+# Same selection for AAA before rbind
+AAA <- AAA %>%
+  select(
+    `Pitcher Name`, `Pitcher ID`, pitch_name, start_speed, spin_rate, extension,
+    IVB, HB, x0, z0
+  )
+
+FSL <- FSL %>%
+  select(
+    `Pitcher Name`, `Pitcher ID`, pitch_name, start_speed, spin_rate, extension,
+    IVB, HB, x0, z0
+  )
+# Helper functions
+calculate_EAA <- function(extension) {
+  extension / 6.3
+}
+
+calculate_SADiff <- function(pfxX, pfxZ, spinDirection) {
+  inSA <- atan2(pfxZ, pfxX) * 180/pi + 90
+  inSA <- ifelse(inSA < 0, inSA + 360, inSA)
+  SADiff <- spinDirection - inSA
+  SADiff <- ifelse(SADiff > 180, SADiff - 360, SADiff)
+  SADiff <- ifelse(SADiff < -180, SADiff + 360, SADiff)
+  return(SADiff)
+}
+
+calculate_VAA <- function(vz0, ay, az, vy0, y0) {
+  -atan((vz0+(az*(-sqrt((vy0*vy0)-(2*ay*(y0-(17/12))))-vy0)/
+                ay))/(-sqrt((vy0*vy0)-(2*ay*(y0-(17/12))))))*(180/pi)
+}
+
+pitcher_summary <- function(game_pk, date) {
+  gdate <- as.Date.character(date)
+  gdate <- as.Date(gdate)
+  tmilb <- mlb_pbp(game_pk)
+  tmilb <- tmilb %>%
+    filter(type == "pitch") %>%
+    select(matchup.batter.fullName, matchup.batter.id, matchup.pitcher.fullName,
+           matchup.pitcher.id, matchup.pitchHand.code, details.type.description,
+           pitchData.startSpeed, pitchData.breaks.spinRate, pitchData.extension,
+           pitchData.coordinates.x0, pitchData.coordinates.y0, pitchData.coordinates.z0,
+           pitchData.coordinates.aX, pitchData.coordinates.aY, pitchData.coordinates.aZ,
+           pitchData.coordinates.vX0, pitchData.coordinates.vZ0, pitchData.coordinates.vY0,
+           pitchData.coordinates.pfxX, pitchData.coordinates.pfxZ,
+           pitchData.breaks.breakVerticalInduced, pitchData.breaks.breakHorizontal,
+           pitchData.breaks.spinDirection)
+  
+  colnames(tmilb) <- c("Batter Name", "Batter ID", "Pitcher Name", "Pitcher ID",
+                       "phand", "pitch_name", "start_speed", "spin_rate", "extension",
+                       "x0", "y0", "z0", "ax", "ay", "az", "vx0", "vz0", "vy0",
+                       "pfxX", "pfxZ", "IVB", "HB", "spinDirection")
+  
+  tmilb <- tmilb %>%
+    mutate(date = gdate)
+  
+  return(tmilb)
+}
+
+calculate_timstuff <- function(game) {
+  game <- game %>%
+    mutate(VAA = calculate_VAA(vz0, ay, az, vy0, y0),
+           EAA = calculate_EAA(extension),
+           SADiff = calculate_SADiff(pfxX, pfxZ, spinDirection),
+           ishandL = ifelse(phand == "L", 1, 0))
+  
+  feature_vars <- c("ishandL", "start_speed", "IVB", "HB", "EAA", "x0", "z0", "spin_rate", "SADiff")
+  complete_rows <- complete.cases(game[, feature_vars])
+  game_complete <- game[complete_rows, ]
+  game_na <- game[!complete_rows,]
+  game_na$TimStuff <- NA
+  
+  game_complete$TimStuff <- scale_TimStuff(
+    predict(model, as.matrix(cbind(game_complete$ishandL, game_complete$start_speed,
+                                   game_complete$IVB, game_complete$HB, game_complete$EAA,
+                                   game_complete$x0, game_complete$z0, game_complete$spin_rate,
+                                   game_complete$SADiff))),
+    -0.002620635, 0.006021368)
+  
+  game_complete <- rbind(game_complete, game_na)
+  return(game_complete)
+}
+
+scale_TimStuff <- function(raw_score, model_mean, model_sd) {
+  scaled_score <- (raw_score - model_mean) / model_sd
+  result <- 100 - (scaled_score * 10)
+  return(result)
+}
+
+summary_table <- function(data) {
+  # Current year summary
+  current_summary <- data %>%
+    group_by(`Pitcher Name`, `Pitcher ID`, pitch_name) %>%
+    summarize(
+      Pitches = n(),
+      'Velo' = round(mean(start_speed, na.rm = TRUE), 1),
+      'Spin' = round(mean(spin_rate, na.rm = TRUE), 0),
+      'Ext' = round(mean(extension, na.rm = TRUE), 1),
+      'IVB' = round(mean(IVB, na.rm = TRUE), 1),
+      'HB' = round(mean(HB, na.rm = TRUE), 1),
+      'RelX' = round(mean(x0, na.rm = TRUE), 1),
+      'RelZ' = round(mean(z0, na.rm = TRUE), 1),
+      'TimStuff' = round(mean(TimStuff, na.rm = TRUE), 0),
+      .groups = "drop"
+    )
+  
+  data_2024 <- rbind(MLB,AAA,FSL) %>%
+    group_by(`Pitcher Name`, `Pitcher ID`, pitch_name) %>%
+    summarize(
+      'Velo24' = round(mean(start_speed, na.rm = TRUE), 1),
+      'Spin24' = round(mean(spin_rate, na.rm = TRUE), 0),
+      'Ext24' = round(mean(extension, na.rm = TRUE), 1),
+      'IVB24' = round(mean(IVB, na.rm = TRUE), 1),
+      'HB24' = round(mean(HB, na.rm = TRUE), 1),
+      'RelX24' = round(mean(x0, na.rm = TRUE), 1),
+      'RelZ24' = round(mean(z0, na.rm = TRUE), 1),
+      .groups = "drop"
+    )
+  
+  # Join and calculate differences
+  combined_data <- current_summary %>%
+    left_join(data_2024, by = c("Pitcher Name", "Pitcher ID", "pitch_name")) %>%
+    mutate(
+      'Velo_Diff' = round(Velo - Velo24, 1),
+      'Spin_Diff' = round(Spin - Spin24, 1),
+      'Ext_Diff' = round(Ext - Ext24, 1),
+      'IVB_Diff' = round(IVB - IVB24, 1),
+      'HB_Diff' = round(HB - HB24, 1),
+      'RelX_Diff' = round(RelX - RelX24, 1),
+      'RelZ_Diff' = round(RelZ - RelZ24, 1)
+    ) %>%
+    arrange(-Pitches)
+  
+  return(combined_data)
+}
+
+# Load TimStuff model
+model <- xgb.load('TimStuff2.model')
+
+# UI Definition
+ui <- fluidPage(
+  titlePanel("MLB/AAA/FSL Pitch Comparison Dashboard"),
+  sidebarLayout(
+    sidebarPanel(
+      width = 2,
+      dateInput("date", "Date:"),
+      selectizeInput("level", "Level:", 
+                     c("MLB", "AAA", "FSL", "College (Statcast Parks Only)", 
+                       "Futures Game", "AFL")),
+      actionButton("submit", "Get Dashboard"),
+      downloadButton("download_summary", "Download Summary")
     ),
-    hr(), # Add a horizontal rule
-    checkboxInput("by_species", "Show species", TRUE),
-    checkboxInput("show_margins", "Show marginal plots", TRUE),
-    checkboxInput("smooth", "Add smoother"),
-  ),
-  plotOutput("scatter")
+    mainPanel(
+      dataTableOutput("schedule")
+    )
+  )
 )
 
+# Server Definition
 server <- function(input, output, session) {
-  subsetted <- reactive({
-    req(input$species)
-    df |> filter(Species %in% input$species)
-  })
-
-  output$scatter <- renderPlot(
-    {
-      p <- ggplot(subsetted(), aes(!!input$xvar, !!input$yvar)) +
-        theme_light() +
-        list(
-          theme(legend.position = "bottom"),
-          if (input$by_species) aes(color = Species),
-          geom_point(),
-          if (input$smooth) geom_smooth()
+  data <- reactiveVal()
+  games <- reactiveVal()
+  summary <- reactiveVal()
+  
+  observeEvent(input$submit, {
+    season <- format(input$date, "%Y")
+    
+    # Get schedule based on selected level
+    schedule_data <- switch(input$level,
+                            "MLB" = mlb_schedule(season = season, level_ids = "1"),
+                            "AAA" = mlb_schedule(season = season, level_ids = "11"),
+                            "FSL" = mlb_schedule(season = season, level_ids = "14"),
+                            "College (Statcast Parks Only)" = mlb_schedule(season = season, level_ids = "22"),
+                            "Futures Game" = mlb_schedule(season = season, level_ids = "21"),
+                            "AFL" = {
+                              sbid <- mlb_schedule(2024, 17) %>%
+                                filter(teams_away_team_name %in% c("Glendale Desert Dogs", "Mesa Solar Sox",
+                                                                   "Peoria Javelinas", "Salt River Rafters",
+                                                                   "Scottsdale Scorpions", "Surprise Saguaros")) %>%
+                                filter(gameday_type == "E")
+                              sbid
+                            }
+    )
+    
+    data(schedule_data)
+    schedule <- data() %>% filter(date == input$date)
+    
+    # Initialize empty games dataframe
+    games_data <- data.frame()
+    
+    # Process each game
+    for(n in 1:nrow(schedule)) {
+      tryCatch({
+        game1 <- pitcher_summary(schedule[n,6], schedule[n,1])
+        games_data <- rbind(game1, games_data)
+      }, error = function(e) {
+        message(paste("Error occurred for game:", schedule[n,6], "on", schedule[n,1]))
+      })
+    }
+    
+    # Calculate TimStuff and create summary
+    games_data <- calculate_timstuff(games_data)
+    games(games_data)
+    summary_data <- summary_table(games_data)
+    summary(summary_data)
+    
+    # Render comparison table
+    output$schedule <- renderDT({
+      datatable(summary_data,
+                options = list(
+                  pageLength = 10,
+                  lengthMenu = c(10, 25, 50, 100),
+                  columnDefs = list(
+                    list(className = 'dt-center', targets = "_all"),
+                    # Names and identifiers
+                    list(width = '150px', targets = c(0, 1)),  # Pitcher Name, Pitcher ID
+                    list(width = '100px', targets = 2),        # pitch_name
+                    list(width = '70px', targets = 3),         # Pitches
+                    # Current year stats
+                    list(width = '50px', targets = c(4:11)),   # Velo, Spin, Ext, IVB, HB, RelX, RelZ, TimStuff
+                    # 2024 stats
+                    list(width = '10px', targets = c(12:18)),  # Velo_2024, Spin_2024, Ext_2024, IVB_2024, HB_2024, RelX_2024, RelZ_2024
+                    # Difference columns
+                    list(width = '50px', targets = c(19:25))   # All _Diff columns
+                  )
+                )
+      ) %>%
+        formatStyle(
+          c('Velo_Diff', 'Spin_Diff', 'Ext_Diff', 'IVB_Diff', 
+            'HB_Diff', 'RelX_Diff', 'RelZ_Diff'),
+          backgroundColor = styleInterval(
+            cuts = 0,
+            values = c('#ffcdd2', '#c8e6c9')
+          )
         )
-
-      if (input$show_margins) {
-        margin_type <- if (input$by_species) "density" else "histogram"
-        p <- p |> ggExtra::ggMarginal(
-          type = margin_type, margins = "both",
-          size = 8, groupColour = input$by_species, groupFill = input$by_species
-        )
-      }
-
-      p
+    })
+    
+  })
+  
+  # Download handler for summary CSV
+  output$download_summary <- downloadHandler(
+    filename = function() {
+      paste("comparison_data_", Sys.Date(), ".csv", sep = "")
     },
-    res = 100
+    content = function(file) {
+      write.csv(summary(), file, row.names = FALSE)
+    }
   )
 }
 
-shinyApp(ui, server)
+shinyApp(ui, server)