Why do RadarCharts graph incorrectly in R? (fmsb)

create_beautiful_radarchart <- function(data, color = "#800000", 
                                        vlabels = colnames(data), vlcex = 0.7,
                                        caxislabels = NULL, title = NULL, ...){
  radarchart(
    data, axistype = 1,
    pcol = color, pfcol = scales::alpha(color, 0.5), plwd = 2, plty = 1,
    cglcol = "grey", cglty = 1, cglwd = 0.8,
    axislabcol = "grey",
    vlcex = vlcex, vlabels = vlabels,
    caxislabels = caxislabels, title = title, ...
  )
}

colors <- c("#00AFBB", "#E0115F", "#800000", "orange")
titles <- c("UnderWeight", "NormalWeight", "OverWeight", "Obese")

for(i in 1:4){
  create_beautiful_radarchart(
    data = df[c(1,2,3,4),], caxislabels = c(5000,10000,15000,20000),
    color = colors[i], title = titles[i],
    seg=3)
}

Why does my code produce radarcharts like:

Here is my entire dataset:

The first two rows of your data frame have to be the maximum and minimum values you want plotted for that series. Since you are filling the areas, you will need to make the colors for the macimum and minimum values NA. You can automate the ranges for each series like this:

range <- as.data.frame(lapply(df, function(x) rev(range(pretty(x)))))

And presumably you want 4 separate radar plots, in which case you can do:

for(i in 1:4){
  create_beautiful_radarchart(
    data = rbind(range, df[i,]), caxislabels = c(5000,10000,15000,20000),
    color = c(NA, NA, colors[i]), title = titles[i], 
    seg=3)
}