Could position nudging be done in an easy proportional way relative to scale as a whole (e.g. using I function)?

[davidhodge931]

It'd be nice to be able to nudge in a proportional way easily. At present, you can calculate your nudge manually in absolute units, but it'd nice if there was a more simple straightforward way.

The I function allows relative positioning when annotating. So was wondering if it'd be possible to support this with nudging? This is adding nudge_x =I(0.05) or position_nudge(x = I(0.05)) to do this, rather than having to manually calculate.

Sorry for all the issues!

library(tidyverse)
library(tidyplots)

#currrently do this 
spendings |> 
  group_by(category) |> 
  summarise(across(amount, sum)) |> 
  ggplot(aes(x = amount, y = category, label = amount)) +
  geom_col() +
  geom_text(nudge_x = 1200 * 0.05) #guessing 1200 is the size of the x scale

  
# # would like to do this
# spendings |> 
#   group_by(category) |> 
#   summarise(across(amount, sum)) |> 
#   ggplot(aes(x = amount, y = category, label = amount)) +
#   geom_col() +
#   geom_text(nudge_x = I(0.05)) 

^{Created on 2025-05-21 with reprex v2.1.1}

[smouksassi]

you can nudge by a fixed amount or proportional to your x value
( you can also do stuff using npc units in ggpp geom_text_npc
what exactly you need to do with your example ?

library(tidyverse)
library(tidyplots)
library(patchwork)
#currrently do this 
a<- spendings |> 
  group_by(category) |> 
  summarise(across(amount, sum)) |> 
  ggplot(aes(x = amount, y = category, label = amount)) +
  geom_col() +
  geom_text(aes(x=amount*1.2)) +
  labs(subtitle = "nudge 20%")
b<- spendings |> 
  group_by(category) |> 
  summarise(across(amount, sum)) |> 
  ggplot(aes(x = amount, y = category, label = amount)) +
  geom_col() +
  geom_text(aes(x=amount+50)) +
  labs(subtitle = "nudge +50 x units")

a | b

^{Created on 2025-05-20 with reprex v2.1.1}

[davidhodge931]

Interesting, thanks @smouksassi I want to be able to easily make the nudging the same for each value, and proportional to the whole of a scale. So multiplying by the aesthetic doesn't work, as large values then get larger nudging. Adding works but it's not proportional without doing some calcs

[teunbrand]

In the development version, since nudge_x has become a 'proper' aesthetic, you can do this: geom_text(aes(nudge_x = after_stat(max(abs(x)) * 0.05))).

[davidhodge931]

That looks pretty cool @teunbrand

However, it would still make the nudge proportional to the max value, rather than proportional to the entire scale size - which is what I want because I often change the limits, expand etc. This code nudges relative to the max value, as I understand it.

I'd say the use of after_stat is still pretty challenging for non-advanced ggplot2 users too.

Is the I solution possible? This would align nicely with how this works in annotating

https://www.tidyverse.org/blog/2024/02/ggplot2-3-5-0/#ignoring-scales

[davidhodge931]

Ideally, I want to be able to add a line of code like this, which would add some space equivalent to that proportion of the scale as a whole...

geom_text(aes(x = xvar + I(0.02)), hjust = 0)

Not sure if it'd be possible..

library(tidyverse)
library(tidyplots)

# can do this based on the max value for where scale min is 0
spendings |>
  group_by(category) |> 
  summarise(across(amount, sum)) |> 
  ggplot(aes(x = amount, y = category, label = amount)) +
  geom_col() +
  geom_text(aes(x = amount + max(amount) * 0.01), hjust = 0) +
  scale_x_continuous(expand = expansion(c(0, 0.1)))

# #prefer to do this based on the scale size, which would work for any limits/expand combo consistently
# spendings |>
#   group_by(category) |> 
#   summarise(across(amount, sum)) |> 
#   ggplot(aes(x = amount, y = category, label = amount)) +
#   geom_col() +
#   geom_text(aes(x = amount + I(0.02)), hjust = 0) +
#   scale_x_continuous(expand = expansion(c(0, 0.1)))

^{Created on 2025-05-22 with reprex v2.1.1}

[teunbrand]

That sounds like you want the same as #5609, in which case, I'll mark this as a duplicate to prune discussion here so it may grow there.