Date created: Thu Aug 23 10:48:36 2018

Here we can visually investigate the trends and patterns in the geographical location, number of Superbown wins and Facebook Likes for each of the 32 NFL teams.Data on Facebook likes was collected by myself from each teams official Facebook site and are correct as of 22/08/18. All NFL team icons were obtained direct from NFL.com. Prior to the analysis this was entered into a local SPSS file and subsequently imported as a data frame into R.

Variable	Info
Team	NFL Team name
Acro	Team 2 or 3 letter acronym
Conference	Denotes whether the team are in the NFC or AFC conference
Latitude	Geographical location of Team stadium
Longitude	Geographical location of Team stadium
SuperbowlWin	Number of Superbowl wins (correct as of August, 2018)
Fanbase	Number of ‘Likes’ on the teams official Facebook page

Below, I’ll be showing you exactly how to create such a plot with the use of basic R programming knowledge and the Leaflet package. If anyone is interested in how I changed the graphics of the R Markdown file, it was done using the prettydoc R package - a walk through guide can be found here (http://yixuan.cos.name/prettydoc/). The tables were also formatted using the lemon package, which can be set to the default table generator using knit_print.data.frame <- lemon_print.

Map Trends

Most popular AFC team is the New England Patriots while the most popular NFC team is the Dallas Cowboys.
Least popular team is the Jacksonville Jaguars.
There are two cities who have two NFL teams.
San Francisco has both the 49ers and the Oakland raiders. The 49ers have both more superbowl wins and more Facebook likes. New York also has both the Giants and the Jets. Similarly, the more popular team, the Giants, have more Facebook likes.
Overall there seems to be a trend that teams with more Superbowl wins have more Facebook likes.
Proportionally, the Seahalks and Bears have lots of Facebook likes given they’ve only won the Superbowl once.
NFL teams are largely spread acorss the North East of the US.

MAP KEY

The NFL team logo is proportional to the number of Superbowl wins with more wins having larger logos. The corresponding team circle is also proportional to the number of Facebook likes each team has and the colour is representative of which NFL conference the team is in.

Load Packages

The following code requires the leaflet package (used to create the interctive map), the haven package to import the data set from an SPSS document and the magrittr package to allow us to use piping (%>%) in R markdown.

library(leaflet) # Interative maps
library(haven) # Import from SPSS
library(magrittr) # Allow piping (%>%) to be used in R markdown

Data set up

Next we want to actually load the data set into thr R environment. As mentioned above, this will be done using the read_sav function from the haven package, as the dataset is coming from an SPSS file. We can also see the top few rows of that data set below.

Stadium_data <- read_sav("~/Documents/GitHub/NFL_Analysis/NFL_winsandlikes.sav")

Team	Acro	Conference	latitude	longitude	SuperbownWin	Fanbase	conferencecolour
Cowboys	DAL	NFC	32.74778	-97.09278	5	8629832	green
Patriots	NE	AFC	42.09092	-71.26435	5	7051620	blue
Steelers	PIT	AFC	40.44679	-80.01576	6	6454093	blue

We also need to set up some additional variables for the icon URL. Due to the consistent URL naming system used by NFL.com, most of the URL for each teams icon are the same with a change in team acroymn. Here we set up the main part of the URL and its suffix to later insert the acroymn between them.

nflurl = c("https://static.nfl.com/static/content/public/static/wildcat/assets/img/logos/teams/")
suffix = c(".svg")

We also eventully want to colour code our markers by NFL conference. Here we create a new coloumn for the data set and assign it blue for all AFC teams and green for all NFC teams.

Stadium_data$conferencecolour <- c("")
Stadium_data$conferencecolour[which(Stadium_data$Conference=='AFC')] <- 'blue'
Stadium_data$conferencecolour[which(Stadium_data$Conference=='NFC')] <- 'green'

Making the Map

Now we come to the fun bit, making the map! This requires a bit more understanding of R coding but I’ll be walking you though it in the comments below so it’ll be more manageable. This code makes use of the piping notation (%>%) which allows multiple functions to be stringed together where the output from the previous function is used as the input for the next and so on.

When we put a marker down for each NFL team, we want to have a different icon type which means we have to loop through all the NFL teams to insert the relevant URL each time.

# Create a variable to contain the map information
# Leaflet and addTiles initiates a viewer and adds the map
my_map <- leaflet() %>% addTiles() 

# Loop through 1:32 which matches the number of NFL teams
for (i in 1:dim(Stadium_data)[1]){
  # Makes marker image - this will be different from each loop
  thisIcon <- makeIcon(
                    # Paste together the URL start and suffix with the team acroymn in the middle
                    iconUrl = paste(nflurl,Stadium_data$Acro[i],suffix, sep=""),
                    # Icon size
                    iconWidth = (31*215/230)+(Stadium_data$SuperbownWin[i])*8,
                    iconHeight = 31+(Stadium_data$SuperbownWin[i])*8,
                    # Which part of the image we want to be on the geographical location
                    # 0,0 corresponds to top left. Here we set it to the middle of the image
                    iconAnchorX = ((31*215/230)+(Stadium_data$SuperbownWin[i])*8)/2,
                    iconAnchorY = (31+(Stadium_data$SuperbownWin[i])*8)/2)

  # Add the Teams marker to the stadiums geographical position
  my_map <- addMarkers(my_map,
                       lat=Stadium_data$latitude[i], 
                       lng= Stadium_data$longitude[i], 
                       # Add a label to the marker for the team name and number of Superbowl wins
                       popup=paste(Stadium_data$Team[i]," : Superbowl Wins ",Stadium_data$SuperbownWin[i],sep=""), 
                       # Set the icon image to be the one we've just created
                       icon=thisIcon)
  # Add a circle second marker at the same location to show facebook fans
  my_map <- addCircles(my_map,
                       lat=Stadium_data$latitude[i], 
                       lng= Stadium_data$longitude[i],
                       # Circumference Line width outside the circle  
                       weight = 1, 
                       # Size of the circle determined by number of fans
                       radius = sqrt((Stadium_data$Fanbase[i]/10)^2),
                       # Add a label to the marker for the team name and number of Facebook likes
                       popup=paste(Stadium_data$Team[i]," : Number Facebook Likes ", Stadium_data$Fanbase[i]), 
                       # Colour the marker by nfl conference
                       color = Stadium_data$conferencecolour[i])
}

# Add a legend to the map for the conference coloured circles
my_map <- addLegend(my_map,labels = c("AFC","NFC"), colors = c("blue","green"))

# Display the Map!
my_map

NFL Team: Location, Wins and Likes

Geographical Location, Superbowl Wins and Facebook likes

Rebecca Elizabeth Kitching

August 2018

Map Trends

MAP KEY

Load Packages

Data set up

Making the Map