How to Create Bar Race Animation Charts

Animated bar race ranking.

You can compare variables over time using animated bar charts by following the bars as if you were betting on a single one.

Let’s create a static graph first, and then one that is dynamic, to compare the two.

Follow the steps below to access the data, or click here to download the post28.xlsx file.

1. https://www.tuik.gov.tr/ (The English language option is available in the upper right corner)

2. Statistics/Transportation and Communication

3. Databases/Road Motor Vehicle Statistics

4. The number of road motor vehicles registered to the traffic/Type of Vehicle&Brand

5. Cars&All, respectively.

6. Monthly (All), Region (Turkey)

library(tidyverse)
library(gganimate)
library(gifski)

df <- readxl::read_excel("data.xlsx") %>% 
  select(-1) %>% 
  slice(-c(1:3)) %>% 
  `colnames<-`(paste0("C",seq(1,ncol(.),1))) %>% 
  mutate(
    C1 = unlist(lapply(str_extract_all(C1, "\\([^()]+\\)"), "[[", 1)),
    C1 = gsub("[()]", "", C1),
    C1 = zoo::na.locf(C1)
  ) %>% 
  pivot_longer(!c(C1,C2), names_to = "month", values_to = "value") %>% 
  mutate(
    month = case_when(
      month == "C3" ~ "01",
      month == "C4" ~ "02",
      month == "C5" ~ "03",
      month == "C6" ~ "04",
      month == "C7" ~ "05",
      month == "C8" ~ "06",
      month == "C9" ~ "07",
      month == "C10" ~ "08",
      month == "C11" ~ "09",
      month == "C12" ~ "10",
      month == "C13" ~ "11",
      month == "C14" ~ "12",
    )
  ) %>% 
  mutate(
    date = as.Date(paste0(C2,"-",month,"-",1)),
    value = as.numeric(value)
  ) %>% 
  rename("brand"="C1") %>% 
  select(date,brand,value) %>% 
  filter(date <= as.Date("2022-07-01")) %>% 
  mutate(value = ifelse(is.na(value), 0, value),
         brand = ifelse(brand == "Diger","Others",brand))

There are 49 different car brands.

df_filtered <- df %>% 
  group_by(date) %>% 
  mutate(p = value / sum(value) * 100) %>% 
  ungroup()

Static plot:

ggplot(df_filtered %>% filter(date == as.Date("2022-07-01")),
       aes(x = reorder(brand,p), y = p, fill = p)) +
  geom_col() +
  geom_text(aes(label = round(p,digits = 1)), hjust = -0.1, size = 5) +
  coord_flip() +
  theme_minimal() +
  theme(legend.position = "none",
        axis.title = element_blank(),
        axis.text.y = element_text(size = 17),
        axis.text.x = element_blank(),
        plot.title = element_text(size = 20, face = "bold"),
        plot.subtitle = element_text(size = 20)) +
  scale_fill_gradient(low = "orange", high = "red") +
  scale_y_continuous(limits = c(0,max(df_filtered$p)+1)) +
  labs(
    title = "Number of Newly Registered Car Brands in Turkey (%)",
    subtitle = "07/2022"
  )

Now it’s time to set the animation up.

df_rank <- df_filtered %>% 
  arrange(date,desc(p)) %>% 
  group_by(date) %>% 
  mutate(rank = row_number(),
         date = format(date,"%Y/%m"),
         p = round(p, digits = 1))

anim <- ggplot(df_rank, aes(x = rank, y = p, fill = p)) +
  geom_col() +
  geom_text(aes(y = 0, label = paste0(brand," ")), hjust = 1) +
  geom_text(aes(y = p, label = p), hjust = -0.1) +
  coord_flip(expand = TRUE) +
  theme_minimal() +
  theme(legend.position = "none",
        axis.title = element_blank(),
        axis.text = element_blank(),
        plot.title = element_text(size = 20, face = "bold"),
        plot.subtitle = element_text(size = 20)) +
  scale_fill_gradient(low = "orange", high = "red") +
  scale_x_reverse() +
  labs(
    title = "Number of Newly Registered Car Brands in Turkey (%)",
    subtitle = "{closest_state}"
  ) +
  transition_states(date) +
  view_follow()

animate(
  anim,
  223,
  fps = 40,
  duration = 50,
  width = 1800,
  height = 1000,
  renderer = gifski_renderer("gganim.gif")
)