Update app.R

app.R

@@ -10,7 +10,11 @@ library(httr)
 library(progressr)
 library(RCurl)
 library(curl)
+library(xgboost)
+library(recipes)
 
+stuffplus_model <- xgb.load("stuffplus_xgb.json")
+stuffplus_recipe <- readRDS("stuffplus_recipe.rds")
 
 # Define columns to remove if they exist
 columns_to_remove <- c(
@@ -215,6 +219,231 @@ merge_with_bat_tracking <- function(csv_data, bat_tracking_data) {
   ))
 }
 
+
+clean_college_data <- function(data, teams = NA){
+  
+  data <- data %>% 
+    mutate(PlayResult = ifelse(PlayResult %in% c("HomeRun", "homerun"), "Homerun", PlayResult), 
+           Batter = sub("(.*),\\s*(.*)", "\\2 \\1", Batter),
+           Pitcher = sub("(.*),\\s*(.*)", "\\2 \\1", Pitcher),
+           Catcher = sub("(.*),\\s*(.*)", "\\2 \\1", Catcher))
+  
+  
+  col <- colnames(data)
+  
+  if ("Top/Bottom" %in% col){ 
+    data <- data %>% 
+      rename(`Top.Bottom` = `Top/Bottom`)
+  }
+  
+  numeric_columns <- c("PitchNo", "PAofInning", "PitchofPA", "PitcherId", "BatterId", "Inning", "Outs", "Balls",
+                       "Strikes", "OutsOnPlay", "RunsScored", "RelSpeed", "VertRelAngle", "HorzRelAngle", "SpinRate", 
+                       "SpinAxis", "RelHeight", "RelSide", "Extension", "VertBreak", "InducedVertBreak", "HorzBreak", 
+                       "PlateLocHeight", "PlateLocSide", "ZoneSpeed", "VertApprAngle", "HorzApprAngle", "ZoneTime", 
+                       "ExitSpeed", "Angle", "Direction", "HitSpinRate", "Distance", "Bearing", "HangTime",
+                       "LastTrackedDistance", "pfxx", "pfxz", "x0", "y0", "z0", "vx0", "vz0", "vy0", "ax0", "ay0", 
+                       "az0",  "EffectiveVelo", "MaxHeight", "SpeedDrop", "ContactPositionX", "ContactPositionY",
+                       "ContactPositionZ", "HomeTeamForeignID", "AwayTeamForeignID", "CatcherId", "ThrowSpeed",
+                       "PopTime", "ExchangeTime", "TimeToBase") 
+  
+  data <- data %>%
+    mutate(across(any_of(numeric_columns), as.numeric),
+           PlateLocHeight = if ("PlateLocHeight" %in% names(.)) 12 * PlateLocHeight else PlateLocHeight,
+           PlateLocSide = if ("PlateLocSide" %in% names(.)) 12 * PlateLocSide else PlateLocSide)
+  
+  
+  data <- data %>% 
+    mutate(TaggedPitchType = case_when(
+      TaggedPitchType == "FourSeamFastBall" ~ "Fastball", 
+      TaggedPitchType %in% c("TwoSeamFastBall", "OneSeamFastBall") ~ "Sinker", 
+      TaggedPitchType == "ChangeUp" ~ "Changeup", 
+      TaggedPitchType == "Undefined" ~ "Other", 
+      T ~ TaggedPitchType
+    ))
+  
+  data <- data %>%
+    mutate(
+      is_csw = case_when(
+        PitchCall %in% c("StrikeSwinging", "StrikeCalled") ~ 1,
+        TRUE ~ 0
+      ),
+      is_swing = case_when(
+        PitchCall %in% c("StrikeSwinging", "FoulBallNotFieldable", "InPlay", 
+                         "FoulBallFieldable", "FoulBall") ~ 1,
+        TRUE ~ 0
+      ),
+      is_whiff = case_when(
+        PitchCall == "StrikeSwinging" & is_swing == 1 ~ 1,
+        PitchCall != "StrikeSwinging" & is_swing == 1 ~ 0,
+        TRUE ~ NA_real_
+      ),
+      in_zone = case_when(
+        PlateLocSide > 9.975 | PlateLocSide < -9.975 |
+          PlateLocHeight  > 40 | PlateLocHeight  < 20 ~ 0,
+        TRUE ~ 1
+      ),
+      chase = case_when(
+        is_swing == 1 & in_zone == 0 ~ 1,
+        is_swing == 0 & in_zone == 0 ~ 0,
+        TRUE ~ NA_real_
+      ),
+      in_zone_whiff = case_when(
+        is_swing == 1 & in_zone == 1 & is_whiff == 1 ~ 1,
+        is_swing == 1 & in_zone == 1 & is_whiff == 0 ~ 0,
+        TRUE ~ NA_real_
+      ),
+      is_hit = case_when(
+        PlayResult %in% c("Single", "Double", "Triple", "Homerun", "HomeRun") & PitchCall == "InPlay" ~ 1,
+        !PlayResult %in% c("Single", "Double", "Triple", "Homerun", "HomeRun") & PitchCall == "InPlay" ~ 0,
+        KorBB == "Strikeout" ~ 0,
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 0,
+        TRUE ~ NA_real_
+      ),
+      slg = case_when(
+        PitchCall == "InPlay" & PlayResult == "Single" ~ 1,
+        PitchCall == "InPlay" & PlayResult == "Double" ~ 2,
+        PitchCall == "InPlay" & PlayResult == "Triple" ~ 3,
+        PitchCall == "InPlay" & PlayResult %in% c("Homerun", "HomeRun") ~ 4,
+        !PlayResult %in% c("Single", "Double", "Triple", "Homerun", "HomeRun") & PitchCall == "InPlay" ~ 0,
+        KorBB == "Strikeout" ~ 0,
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 0,
+        TRUE ~ NA_real_
+      ),
+      on_base = case_when(
+        PitchCall == "InPlay" & PlayResult %in% c("Single", "Double", "Triple", "Homerun", "HomeRun") ~ 1,
+        PitchCall %in% c("HitByPitch") | KorBB == "Walk" ~ 1,
+        PitchCall == "InPlay" & PlayResult %in% c("Out", "Error", "FieldersChoice") & PlayResult != "Sacrifice" ~ 0,
+        KorBB == "Strikeout" ~ 0,
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 0,
+        TRUE ~ NA_real_
+      ),
+      is_hard_hit = case_when(
+        ExitSpeed >= 95 & PitchCall == "InPlay" ~ 1,
+        ExitSpeed < 95 & PitchCall == "InPlay" ~ 0,
+        TRUE ~ NA_real_
+      ),
+      woba = case_when(
+        PitchCall == "InPlay" & PlayResult == "Single" ~ 0.95,
+        PitchCall == "InPlay" & PlayResult == "Double" ~ 1.24,
+        PitchCall == "InPlay" & PlayResult == "Triple" ~ 1.47,
+        PitchCall == "InPlay" & PlayResult %in% c("Homerun", "HomeRun") ~ 1.71,
+        KorBB == "Walk" ~ 0.82,
+        PitchCall %in% c("HitByPitch") ~ 0.85,
+        KorBB == "Strikeout" ~ 0,
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 0,
+        PitchCall == "InPlay" & !PlayResult  %in% c("Single", "Double" ,"Triple" ,"Homerun", "HomeRun") ~ 0,
+        TRUE ~ NA_real_
+      ),
+      wobacon = case_when(
+        PitchCall == "InPlay" & PlayResult == "Single" ~ 0.95,
+        PitchCall == "InPlay" & PlayResult == "Double" ~ 1.24,
+        PitchCall == "InPlay" & PlayResult == "Triple" ~ 1.47,
+        PitchCall == "InPlay" & PlayResult %in% c("Homerun", "HomeRun") ~ 1.71,
+        PitchCall == "InPlay" & !PlayResult  %in% c("Single", "Double" ,"Triple" ,"Homerun", "HomeRun") ~ 0,
+        TRUE ~ NA_real_
+      ),
+      is_plate_appearance = ifelse(
+        PitchCall %in% c("InPlay", "HitByPitch") | KorBB %in% c("Strikeout", "Walk") | PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking"), 1, 0
+      ),
+      is_at_bat = case_when(
+        PitchCall == "InPlay" & !PlayResult %in% c("StolenBase", "Sacrifice", "CaughtStealing", "Undefined") ~ 1,
+        KorBB == "Strikeout" ~ 1,
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 1,
+        TRUE ~ 0
+      ),
+      is_walk = case_when(
+        is_plate_appearance == 1 & KorBB == "Walk" ~ 1,
+        is_plate_appearance == 1 & KorBB != "Walk" ~ 0,
+        TRUE ~ NA_real_
+      ),
+      is_k = case_when(
+        is_at_bat == 1 & KorBB == "Strikeout" ~ 1,
+        is_at_bat == 1 & KorBB != "Strikeout" ~ 0, 
+        PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 1,
+        TRUE ~ NA_real_
+      ), 
+      is_put_away = case_when(
+        Strikes == 2 & KorBB == "Strikeout" ~ 1,
+        Strikes == 2 & KorBB != "Strikeout" ~ 0,
+        Strikes == 2 & PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking") ~ 1,
+        TRUE ~ NA_real_
+      ), 
+      OutsOnPlay = ifelse(KorBB == "Strikeout" | PlayResult %in% c("StrikeoutSwinging", "StrikeoutLooking"), OutsOnPlay + 1, OutsOnPlay)
+    )
+  
+  data <- data %>% 
+    mutate(event_type = case_when(
+      PitchCall %in% c("StrikeSwinging", "StrkeSwinging") ~ "Whiff", 
+      PitchCall %in% c("StriekC", "StrikeCalled") ~ "Called Strike", 
+      PitchCall %in% c("FoulBallFieldable", "FoulBall", "FoulBallNotFieldable",
+                       "FouldBallNotFieldable") ~ "Foul Ball", 
+      PitchCall %in% c("BallCalled", "BallinDirt", "BallIntentional", "BalIntentional") ~ "Ball", 
+      PitchCall == "HitByPitch" ~ "HBP", 
+      PitchCall == "InPlay" & PlayResult %in% c("Out", "FieldersChoice",
+                                                "Error", "error", 
+                                                "Sacrifice") ~ "Field Out", 
+      PitchCall == "InPlay" & PlayResult == "Single" ~ "Single", 
+      PitchCall == "InPlay" & PlayResult == "Double" ~ "Double",
+      PitchCall == "InPlay" & PlayResult == "Triple" ~ "Triple",
+      PitchCall == "InPlay" & PlayResult == "Homerun" ~ "Home Run", 
+      T ~ NA
+    ))
+  
+  
+  return(data)
+  
+  
+}
+
+
+
+predict_stuffplus <- function(data) {
+
+  predict_data <- data %>% 
+    mutate(RelSide = case_when(
+      PitcherThrows == "Right" ~ RelSide,
+      PitcherThrows == "Left" ~ -RelSide,
+      PitcherThrows %in% c("Both", "Undefined") & RelSide > 0 ~ RelSide,
+      PitcherThrows %in% c("Both", "Undefined") & RelSide < 0 ~ -RelSide),
+          ax0 = case_when(
+            PitcherThrows == "Right" ~ ax0,
+            PitcherThrows == "Left" ~ -ax0,
+            PitcherThrows %in% c("Both", "Undefined") & ax0 > 0 ~ ax0,
+            PitcherThrows %in% c("Both", "Undefined") & ax0 < 0 ~ -ax0),
+                 PlateLocHeight = PlateLocHeight*12,
+                 PlateLocSide = PlateLocSide*12,
+          ax0 = -ax0) %>%
+    group_by(Pitcher, GameID) %>%
+    mutate(
+      primary_pitch = case_when(
+        any(TaggedPitchType == "Fastball") ~ "Fastball",
+        any(TaggedPitchType == "Sinker") ~ "Sinker",
+        TRUE ~ names(sort(table(TaggedPitchType), decreasing = TRUE))[1]
+      )
+    ) %>%
+    group_by(Pitcher, GameID, primary_pitch) %>%
+    mutate(
+      primary_az0 = mean(az0[TaggedPitchType == primary_pitch], na.rm = TRUE),
+      primary_velo = mean(RelSpeed[TaggedPitchType == primary_pitch], na.rm = TRUE)
+    ) %>%
+    ungroup() %>% 
+    mutate(az0_diff = az0 - primary_az0, 
+           velo_diff = RelSpeed - primary_velo)
+
+
+  df_processed <- bake(stuffplus_recipe, new_data = predict_data)
+
+  df_matrix <- as.matrix(df_processed)
+                                 
+  raw_stuff <- predict(stuffplus_model, df_matrix)
+
+  data$raw_stuff <- raw_stuff
+
+  return(data)
+  
+}
+                                            
+
                                             
                                             
 # UI
@@ -1652,8 +1881,18 @@ observe({
           NULL
         }
       }, error = function(e) { NULL })
-      
       if (!is.null(data) && nrow(data) > 0) {
+        scrape_status_msg("Processing data...")
+        
+        data <- tryCatch({
+          d <- clean_college_data(data)
+          d <- predict_stuffplus(d)
+          d
+        }, error = function(e) {
+          scrape_status_msg(paste("Processing error:", e$message))
+          data  # return unprocessed data if it fails
+        })
+        
         scraped_data(data)
         scrape_status_msg(paste0("Done! ", nrow(data), " rows × ", ncol(data), " columns."))
       } else {