Webscraping from CNN.com using rvest

I am trying to webscrape the exit poll data from CNN.com using rvest but can't seem to figure it out.

I am doing this:

url <- read_html("http://www.cnn.com/election/results/exit-polls/arizona/president")
nodes <- html_nodes(url, ".exit-poll_table")

I would like to get all the table values and store them as data frames but I am getting null values for my nodes variable. Any help is appreciated. Thank you