When data visualization goes beautifully wrong. @accidental__aRt Tweet Curated by @kara_woo and @ErikaMudrak

I call this “world’s tallest spiral staircase.” Intended to make a binary pie chart but forgot to lump all those little guys together first.

via @JennyBryan

Learning how to use `geom_crossbar()`: `ggplot(mtcars, aes(x=cyl, y=mpg, fill=gear)) + geom_crossbar(aes(ymin=mpg-2, ymax=mpg+2))`. This is not what I was looking for.

via Drew

Screwed up trying to make a triangle mesh based on an image of leaves and ended up with this instead.

via Chris

I inserted some code to plot vectors and fits to those vectors to help debug. It produced this strangely pleasing mess.

via Noah Silbert

World TB drug resistance patterns using rworldmap…

via Pete

Made in NetworkX in Python. It ought to be a social network graph.

via Tim

Experimenting with visualizing several dimensions and measures in small multiples. geordielad

Contour plot of random repetitive noise

via @ErikaMudrak

Mistaking ‘cex’ for ‘pch’ leads to a nice example of minimalism.

via iv2sls

I call this one “Every upward trend contains a few downward trends screaming to be let out.” via @infotroph