Graphing the birthplaces of the 2016 MLB players

mlb_birthplaces.R

# Interesting table that presents the number of 2016 MLB players 
# from each of 50 US States.

# http://www.baseball-almanac.com/players/birthplace.php?y=2016

# Interested in mapping the data

# Use XML package to read in data

library(XML)
d <- readHTMLTable("http://www.baseball-almanac.com/players/birthplace.php?y=2016")

# put these data in a data frame

# first convert to character type and make the values into a 
# long vector of 50 elements

d1 <- d[[1]][14:30, 2:4]
for (j in 1:3)
  d1[, j] <- as.character(d1[, j])
d1 <- unlist(d1)

# extract the numeric part of the string

library(stringr)
values <- as.numeric(str_extract(d1, '[0-9]+'))

# extract the state names

state_names <- str_extract(d1, '[a-zA-z ]+')

# remove DC from each vector

values <- values[-48]
state_names <- state_names[-48]

# create the data frame

final <- data.frame(State=state_names, 
                    N_Player=as.numeric(values))

# also need populations of all states
# here I try the htmltab package to extract the data information

library(htmltab)
p <- htmltab("https://en.wikipedia.org/wiki/List_of_U.S._states_and_territories_by_population")

# choose the rows that just correspond to states
# remove commas from population numbers


p <- p[c(1:29, 31:49, 51:52), ]
p <- p[, 3:4]
names(p) <- c("State", "Population")
p$Population <- as.numeric(str_replace_all(p$Population, ",", ""))

# merge the population data with the MLB data

library(dplyr)
final_data <- inner_join(final, p)

# compute percent MLB

final_data <- mutate(final_data, 
                     Pct_MLB = 100 * N_Player / Population,
                     state=tolower(State))

# do the mapping

library(mapdata)
states <- map_data("state")
states <- inner_join(states, final_data, by=c("region"="state"))

library(ggplot2)
TH <- theme(
  plot.title = element_text(
    colour = "red",
    size = 18,
    hjust = 0.5,
    vjust = 0.8,
    angle = 0
  )
)
ggplot(data = states) + 
  geom_polygon(aes(x = long, y = lat, 
          fill = Pct_MLB, group = group), color = "white") + 
  coord_fixed(1.3) +
  scale_fill_gradient(low = "yellow", high = "red") +
  ggtitle("Density of 2016 MLB Players by State") + TH