Asymmetrical dot positioning using geom_dotplot

I just fixed an issue I was having with geom_dotplot and have run into another. Namely that some of the levels have dots that are off centre from the x-axis tick marks which, whilst not a huge problem, makes my plot look a lot messier than it should.

my data is:

structure(list(ID = structure(c(14L, 16L, 2L, 2L, 2L, 3L, 3L, 
3L, 3L, 6L, 6L, 6L, 6L, 6L, 6L, 7L, 10L, 10L, 13L, 14L, 15L, 
16L, 16L, 19L), levels = c("A", "B", "C", "D", "E", "F", "G", 
"H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T"
), class = "factor"), value = c(2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 
2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2.5, 2, 2.5, 2, 2.5, 1, 2.5, 2.5, 
2.5, 2.5, 2.5, 2.5)), class = c("grouped_df", "tbl_df", "tbl", 
"data.frame"), row.names = c(NA, -24L), groups = structure(list(
    ID = structure(c(2L, 3L, 6L, 7L, 10L, 13L, 14L, 15L, 16L, 
    19L), levels = c("A", "B", "C", "D", "E", "F", "G", "H", 
    "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T"
    ), class = "factor"), .rows = structure(list(3:5, 6:9, 10:15, 
        16L, 17:18, 19L, c(1L, 20L), 21L, c(2L, 22L, 23L), 24L), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -10L), .drop = TRUE))

my code is:

ggplot(data, aes(x = ID, fill=as.factor(value))) +
  geom_dotplot(dotsize = 1, stackratio = 1.2, 
               method = "histodot", 
               stackgroups = TRUE) +
  labs(title = "title", 
       x = "", 
       y = "",
       fill = "value") +
  theme_bw() +
  scale_y_continuous(NULL, breaks = NULL) + 
  scale_x_discrete(drop = FALSE) +
  theme(plot.title = element_text(size= 13), 
        axis.text.x = element_text(size = 10),
        plot.margin = unit(c(0, 0.2, 0, 0.2), "inches"))

Here is an image of the plot:

I think that when your x-axis is discrete, a "histodot" may not be appropriate (histograms are only meaningful for numeric data). In that case, use the default method for binning (dotdensity).

ggplot(df1, aes(x=ID, fill=factor(value), group=value)) + 
  geom_dotplot(binpositions = "all",
               dotsize = 1, stackratio = 1.2, 
               stackgroups = TRUE) + ...