lines_vs_anim_bars.R

library(babynames)
library(tidyverse)
library(ggrepel)

dat <- babynames %>%
  filter(sex=='M') %>%
  mutate(chars = nchar(name),
         last_char = substr(name, chars, chars)) %>%
  group_by(year, last_char) %>%
  summarise(prop = sum(prop))

label_dat <- 
  dat %>%
  group_by(last_char) %>%
  slice(which.max(prop))

ggplot(dat, aes(x=year, y=prop*100, color=last_char)) +
  geom_line() +
  geom_point(data=label_dat) +
  geom_label_repel(data= filter(label_dat, prop*100 >= 4),
                   aes(label=toupper(last_char)),
                   #fill='gray90', color='black',
                   size=3, segment.size = NA) +
  theme_light() +  
  theme(legend.position='none') +
  labs(title = 'Distribution of last letters of US male names',
       subtitle = '(points show high water mark for each letter)',
       y = 'Percent of babies born that year')