Plotting Air Yards Auto-correlation

air_yards_plot.R

library(dplyr)
library(ggplot2)
library(rvest)
library(tidyr)

html.doc <- read_html('http://www.footballoutsiders.com/stat-analysis/2016/quarterbacks-and-progression-air-yards')

# Extract table
raw.table <- html.doc %>%
  html_table() %>%
  first

# Get the actual table
my.data <- raw.table %>% tail(-2)

# Set the column names
colnames(my.data) <- raw.table[2,]

# Clean up the table and convert to long format.
cleaned <- my.data %>%
  gather(year, air.yards, -Quarterback) %>%
  filter(year != 'AVG', air.yards != '-') %>%
  mutate(year = as.numeric(year),
         air.yards = as.numeric(air.yards))

# Density of Air yards
cleaned %>%
  ggplot(aes(x = air.yards)) +
  geom_density()

# Self-join to get scatter plot
cleaned %>%
  mutate(year = (year - 1)) %>%
  rename(air.yards2 = air.yards) %>%
  inner_join(cleaned) %>%
  ggplot(aes(x = air.yards, y = air.yards2)) +
  geom_point() +
  geom_smooth(method = 'lm') +
  xlab('Avg. Air Yards Previous Year') +
  ylab('Avg. Air Yards')