Animated sorted bar chart: problem with overlapping bars

Question

I created an animated bar chart which displays the scored goals by some players. Below the whole code is displayed how I came to the output.

The animation works as wishe

Answer 1

Edited solution based on clarification:

gap %>%

  # for each player, note his the rank from his previous day
  group_by(Player) %>%
  arrange(Gameday) %>%
  mutate(prev.rank = lag(rank)) %>%
  ungroup() %>%

  # for every game day,
  # sort players by rank & break ties by previous day's rank
  group_by(Gameday) %>%
  arrange(rank, prev.rank) %>%
  mutate(x = seq(1, n())) %>%
  ungroup() %>%

  ggplot(aes(x = x, y = Goals, fill = Player, color = Player)) +
  # geom_tile(aes(y = Goals/2, height = Goals, width = width)) +
  geom_col() +
  geom_text(aes(y = 0, label = Player), hjust = 1) +
  geom_text(aes(label = Value_lbl), hjust = 0) +

  # rest of the code below is unchanged from the question
  coord_flip(clip = "off", expand = FALSE) +
  scale_y_continuous(labels = scales::comma) +
  scale_x_reverse() +
  guides(color = FALSE, fill = FALSE) +
  labs(title = "Gameday {closest_state}", x="", y = "Goals scored") +
  theme(plot.title = element_text(hjust = 0, size = 22),
        axis.ticks.y = element_blank(), 
        axis.text.y  = element_blank(),
        plot.margin = margin(1,1,1,4, "cm")) +
  transition_states(Gameday, transition_length = 4, state_length = 1) +
  ease_aes('cubic-in-out')

Original solution:

gap %>%

  # for each player, note his the rank from his previous day
  group_by(Player) %>%
  arrange(Gameday) %>%
  mutate(prev.rank = lag(rank)) %>%
  ungroup() %>%

  # for every game day & every rank,
  # reduce tile width if there are multiple players sharing that rank, 
  # sort players in order of who reached that rank first, 
  # & calculate the appropriate tile midpoint depending on how many players are there
  group_by(Gameday, rank) %>%
  mutate(n = n_distinct(Player)) %>%
  mutate(width = 0.9 / n_distinct(Player)) %>%
  arrange(prev.rank) %>%
  mutate(x = rank + 0.9 * (seq(1, 2 * n() - 1, by = 2) / 2 / n() - 0.5)) %>%
  ungroup() %>%

  ggplot(aes(x = x, fill = Player, color = Player)) +
  geom_tile(aes(y = Goals/2, height = Goals, width = width)) +
  geom_text(aes(y = 0, label = Player), hjust = 1) +
  geom_text(aes(y = Goals, label = Value_lbl), hjust = 0) +

  # rest of the code below is unchanged from the question
  coord_flip(clip = "off", expand = FALSE) +
  scale_y_continuous(labels = scales::comma) +
  scale_x_reverse() +
  guides(color = FALSE, fill = FALSE) +
  labs(title = "Gameday {closest_state}", x="", y = "Goals scored") +
  theme(plot.title = element_text(hjust = 0, size = 22),
        axis.ticks.y = element_blank(), 
        axis.text.y  = element_blank(),
        plot.margin = margin(1,1,1,4, "cm")) +
  transition_states(Gameday, transition_length = 4, state_length = 1) +
  ease_aes('cubic-in-out')

Note: This isn't perfect. I imagine the simple logic above for determining player order within the same day / rank won't be ideal if there are too many players / too many days, since it only looks backwards by one day. But it works for this example, & I don't know enough about football (at least I think this is football?) to extrapolate about your use case.