cleanData.Rmd

---
title: "cleanData"
author: "Montgomery Inman"
date: "2024-08-27"
output: html_document
---

```{r}
library(cfbfastR)
library(cfbplotR)
library(tidyverse)
library(gt)
library(gtExtras)
```

Pull Data:

```{r}
team = "Iowa State"
year = 2024
week = 1

#Getting PBP data
pbpForEpa = cfbd_pbp_data(year = year, team = team, week = week, epa_wpa = T) %>%
  filter(down > 0,
         (rush == 1 | pass == 1),
         !str_detect(play_text, "Kneel"))

#Getting gameid for the GAME
gameid = pbpForEpa$game_id[1]

#Getting advanced game log data for the GAME
advancedBox = cfbd_game_box_advanced(gameid, long = F) 

#Getting team stats for the GAME
teamStats = cfbd_game_team_stats(year = year, game_id = gameid)

#Separating teamStats completions and attempts
teamStats = teamStats %>% separate_wider_delim(completion_attempts, delim = "-", names = c("completions", "attempts"))

#Separating teamStats third down efficiency attempts
teamStats = teamStats %>% separate_wider_delim(third_down_eff, delim = "-", names = c("thirdDownCompleted", "thirdDownAttempts"))
```

pbpForEpa: gameid, home & away teams
advancedBox: 
teamStats: points


```{r}
#Seperating the team stats df with home and away
homeTeamStats = teamStats %>% filter(home_away == "home")
awayTeamStats = teamStats %>% filter(home_away == "away")

#Seperating the advanced box df with home and away
homeAdvBox = advancedBox %>% filter(team == pbpForEpa$home[1])
awayAdvBox = advancedBox %>% filter(team == pbpForEpa$away[1])

#Seperating the pbpForEpa df with home and away
homePBP = pbpForEpa %>%
  filter(pos_team == pbpForEpa$home[1])
awayPBP = pbpForEpa %>%
  filter(pos_team == pbpForEpa$away[1])

#Creating home Pbp for passes
homePBPpass = homePBP %>%
  filter(pass == 1)

#Creating away Pbp for passes
awayPBPpass = awayPBP %>%
  filter(pass == 1)

#Creating home Pbp for rushes
homePBPrush = homePBP %>%
  filter(rush == 1)

#Creating away Pbp for rushes
awayPBPrush = awayPBP %>%
  filter(rush == 1)

#Calculating EPA per PLAY for home & away
homeEPAperPlay = sum(homePBP$EPA, na.rm = T) / nrow(homePBP)
awayEPAperPlay = sum(awayPBP$EPA, na.rm = T) / nrow(awayPBP)

#Calculating EPA per PASS for home & away
homeEPAperPass = sum(homePBPpass$EPA, na.rm = T) / nrow(homePBPpass)
awayEPAperPass = sum(awayPBPpass$EPA, na.rm = T) / nrow(awayPBPpass)

#Calculating EPA per RUSH for home & away
homeEPAperRush = sum(homePBPrush$EPA, na.rm = T) / nrow(homePBPrush)
awayEPAperRush = sum(awayPBPrush$EPA, na.rm = T) / nrow(awayPBPrush)
```


```{r}
#Creating the data frame for the gt table
dfForGt = data.frame(
  "Stat" = c("Score",
             "EPA / Play",
             "Yards / Play",
             "Success Rate",
             "EPA / Pass",
             "Yards / Completion",
             "Completion %",
             "Attempts",
             "Yards",
             "EPA / Rush",
             "Yards / Rush",
             "Stuff Rate",
             "Attempts",
             "Yards",
             "Third Down %"
  ),
  home = c(homeTeamStats$points,
           homeEPAperPlay,
             as.numeric(homeTeamStats$total_yards) / homeAdvBox$ppa_plays, 
             homeAdvBox$success_rates_overall_total,
             homeEPAperPass,
             as.numeric(homeTeamStats$net_passing_yards) / as.numeric(homeTeamStats$completions),
             as.numeric(homeTeamStats$completions) / as.numeric(homeTeamStats$attempts),
             homeTeamStats$attempts,
             homeTeamStats$net_passing_yards,
             homeEPAperRush,
             homeTeamStats$yards_per_rush_attempt,
             homeAdvBox$rushing_stuff_rate,
             homeTeamStats$rushing_attempts,
             homeTeamStats$rushing_yards,
             as.numeric(homeTeamStats$thirdDownCompleted) / as.numeric(homeTeamStats$thirdDownAttempts)),
  away = c(awayTeamStats$points,
           awayEPAperPlay,
             as.numeric(awayTeamStats$total_yards) / awayAdvBox$ppa_plays,
             awayAdvBox$success_rates_overall_total,
             awayEPAperPass,
             as.numeric(awayTeamStats$net_passing_yards) / as.numeric(awayTeamStats$completions),
             as.numeric(awayTeamStats$completions) / as.numeric(awayTeamStats$attempts),
             awayTeamStats$attempts,
             awayTeamStats$net_passing_yards,
             awayEPAperRush,
             awayTeamStats$yards_per_rush_attempt,
             awayAdvBox$rushing_stuff_rate,
             awayTeamStats$rushing_attempts,
             awayTeamStats$rushing_yards,
             as.numeric(awayTeamStats$thirdDownCompleted) / as.numeric(awayTeamStats$thirdDownAttempts))
)

#Replacing the columns with the home and away teams
colnames(dfForGt)[2] <- pbpForEpa$home[1]
colnames(dfForGt)[3] <- pbpForEpa$away[1]

#Setting each column to be numeric 
dfForGt[,2] = as.numeric(dfForGt[,2])
dfForGt[,3] = as.numeric(dfForGt[,3])
```


```{r}
#Getting data for ALL team stats for 2023 season
combined_data_team_stats <- data.frame()
weeksForPercentiles = c(1:15)

for (weeks in weeksForPercentiles) {
  temp = cfbd_game_team_stats(
  year = 2023,
  week = weeks,
  season_type = "regular",
  team = NULL,
  conference = NULL,
  game_id = NULL,
  rows_per_team = 1
  )
  
  combined_data_team_stats <- rbind(combined_data_team_stats, temp)
}

#Making combined_data_team_stats into a data frame 
combined_data_team_stats = as.data.frame(combined_data_team_stats)

#Getting data for ALL adv stats for 2023 season
combined_data_adv_stats <- data.frame()
weeksForPercentiles = c(1:15)
weekData = data.frame()

#Have to get ALL the game ids 
for (weeks in weeksForPercentiles) {
  temp = cfbd_game_info(
    year = year,
    week = weeks,
    season_type = "regular",
    division = "fbs",
  )
  
  weekData = rbind(weekData, temp)
  
  gameIds = weekData$game_id
}

#for all the gameids, get ALL the advanced stats for the 2023 season
combined_data_adv_stats <- do.call(rbind, lapply(gameIds, function(game) {
  cfbd_game_box_advanced(game_id = game)
}))

#Getting ALL play by play data for the 2023 season
pbp2023 = cfbd_pbp_data(year = year, epa_wpa = T) %>%
  filter(down > 0,
         !str_detect(play_text, "Kneel"),
         (rush == 1 | pass == 1))
```

```{r}
combined_data_team_stats = combined_data_team_stats %>% separate_wider_delim(completion_attempts, delim = "-", names = c("completions", "attempts")) 

combined_data_team_stats = combined_data_team_stats %>% separate_wider_delim(third_down_eff, delim = "-", names = c("thirdDownCompleted", "thirdDownAttempts"))

combined_data_team_stats$completionPercent = as.numeric(combined_data_team_stats$completions) / as.numeric(combined_data_team_stats$attempts)
```



```{r}
#EFFICIENCY
#"EPA / Play", pbp2023
#"Yards / Play", 
#"Success Rate",
#"Third Down %"

#PASSING
#"EPA / Pass",
#"Yards / Completion",
#"Completion %",
#"Attempts",
#"Yards",

#RUSHING
#"EPA / Rush",
#"Yards / Rush", combined_data_team_stats
#"Stuff Rate", combined_data_adv_stats
#"Attempts",
#"Yards",

```