How to sort a horizontal bar-chart in descending order with ggplot?

Question

I´m trying to create a plot that shows all main categories and their summed up pledged amount in million USD. Therefore I took a Kickstarter data frame, put the pledged amount usd_pledged as x-value and the category main_category as the y-value.

To make it more appealing I used fill = main_category. After that some labelling took place labs(),theme_classic(), theme(),guides() and scale_x_continous()

Here is my code:

ggplot(Kickstarter, aes(usd_pledged/1000000,                               
                            main_category,                                   
                            fill = main_category)) +                               
  geom_histogram(stat = "identity") +
  labs(x="pledged amount in mio USD", y = "main categories") +
  theme_classic() +
  theme(legend.position="right") +
  guides(fill=guide_legend(title="categories")) +
  scale_x_continuous(expand = c(0,0))

The result ist: basic-plot_Kickstarter

Now I´d like to plot this graph in a descending order. I tried using reorder() to get a descending order:

ggplot(Kickstarter, aes(usd_pledged/1000000,                               
                            reorder(main_category, +usd_pledged),                                   
                            fill = main_category)) +                               
  geom_histogram(stat = "identity") +
  labs(x="pledged amount in mio USD", y = "main categories") +
  theme_classic() +
  theme(legend.position="right") +
  guides(fill=guide_legend(title="categories")) +
  scale_x_continuous(expand = c(0,0))

which only kind of sorts it, as the result is: halfway sorted plot_Kickstarter

Is there a way to order it correctly? (Games->Design->Technology->Film&Video->Music->Publishing->Food->Fashion->Art->Comics->Theater->Photography->Dance->Journalism->Crafts)

If you need more info, just let me know. This is my first question here :)

head()

# A tibble: 6 x 20
          ID name  category main_category currency deadline    goal
       <dbl> <chr> <chr>    <chr>         <chr>    <date>     <dbl>
1 1000003930 Gree~ Narrati~ Film & Video  USD      2017-11-01 30000
2 1000004038 Wher~ Narrati~ Film & Video  USD      2013-02-26 45000
3 1000007540 Tosh~ Music    Music         USD      2012-04-16  5000
4 1000014025 Mona~ Restaur~ Food          USD      2016-04-01 50000
5 1000030581 Chas~ Drinks   Food          USD      2016-03-17 25000
6  100004721 Of J~ Nonfict~ Publishing    CAD      2013-10-09  2500
# ... with 13 more variables: launched <chr>, pledged <dbl>,
#   state <chr>, backers <dbl>, country <chr>, usd_pledged <dbl>,
#   usd_pledged_real <dbl>, usd_goal_real <dbl>, percent <dbl>,
#   launch_year <chr>, runtime <dbl>, state_dv <dbl>, Group <chr>

UPDATE

After trying out

    Kickstarter %>% 
mutate(main_category = fct_reorder(main_category, usd_pledged)) %>% 
ggplot(aes(x=main_category, y=usd_pledged, fill=main_category)) +
      geom_col()+
      coord_flip()

it still remains in chaos, as you can see here:

Answer 1

Update: I added some fake values for usd_pledged since the column is not reconstructible:

library(forcats)  
library(ggplot2)
library(dplyr)
  

df %>% 
  mutate(main_category = fct_reorder(main_category, usd_pledged)) %>% 
  ggplot(aes(x=main_category, y=usd_pledged, fill=main_category)) +
  geom_col()+
  coord_flip()

First answer: I am not sure but I think you need geom_col or geom_bar Here is an example with the mtcars dataset:

1. prepare mtcars to get cars column and mpg
2. use fct_reorder from forcats to order cars by mpg
3. use ggplot with geom_col
4. finally coord_flip()

library(forcats)  
library(ggplot2)
library(dplyr)
  
mtcars %>% 
  rownames_to_column("cars") %>% 
  select(cars, mpg) %>% 
  mutate(cars = fct_reorder(cars, mpg)) %>% 
  ggplot(aes(cars, mpg, fill=cars)) +
  geom_col()+
  coord_flip()