CFB EDA.Rmd

---
title: "CFB EDA"
author: "Seth Goldman"
date: "August 18, 2019"
output: html_document
---

```{r setup, include=FALSE}
library(dplyr)
library(ggplot2)
library(inspectdf)
library(forcats)
```

### Data acquisition
Data is queried from the `CFB Data Acquisition.R`

```{r read_data}
all_plays_2018 <- readRDS(file = "all_plays_2018.rds")
all_drives_2018 <- readRDS(file = "all_drives_2018.rds")
all_games_2018 <- readRDS(file = "all_games_2018.rds")

```

```{r preview_data}

glimpse(all_games_2018)

glimpse(all_drives_2018)

glimpse(all_plays_2018)

```

```{r explore_games}

all_games_2018 %>% 
  ggplot(aes(x=home_points,y=away_points, color = week))+
  geom_point()

all_games_2018 %>% 
  ggplot(aes(conference_game,(home_points+away_points)))+
  geom_boxplot()+
  coord_flip()+
  ylab("total_points")+
  ggtitle("total points in conference vs non-conf games")


all_games_2018 %>% 
  filter(conference_game == TRUE) %>% 
  ggplot(aes(fct_reorder(
              home_conference,
              (home_points+away_points)
              ),
             (home_points+away_points)))+
  geom_boxplot()+
  coord_flip()+
  ylab("total_points")+
  xlab("conference")+
  ggtitle("total points by conference",
            subtitle = "conference games only")
```


```{r explore_drives}

all_drives_2018 %>% 
  group_by(offense) %>% 
  summarise("total_drives" = n_distinct(id),
            "total_plays" = sum(plays)) %>% 
  arrange(desc(total_drives))

all_drives_2018 %>%
  count(drive_result) %>% 
  ggplot(aes(x=reorder(drive_result,n), y = n))+
  geom_bar(stat="identity")+
  coord_flip()+
  ggtitle("drive result types")+
  xlab("")+
  ylab("observations")


all_drives_2018 %>% 
  count(elapsed.minutes,elapsed.seconds)

all_drives_2018 %>% 
  filter(drive_result %in% c("PUNT","TD", "FG", "INT","DOWNS","FUMBLE","MISSED FG")) %>% 
  filter(elapsed.minutes > 0) %>% 
  ggplot(aes(x=fct_reorder(drive_result,elapsed.minutes),y=elapsed.minutes))+
  geom_boxplot()+
  coord_flip()+
  ggtitle("drive minutes by result type")+
  xlab("")+
  ylab("drive time elapsed (whole mins)")
```


```{r explore_plays}
all_plays_2018 %>% 
  count(play_type) %>% 
  ggplot(aes(x=fct_reorder(play_type,n), y = n))+
  geom_bar(stat = "identity")+
  coord_flip()

```