• If the “story” is simple, keep it simple.
• If the “story” is complex, make it look simple.
• Tell the truth – don’t distort the data.

– Ross Ihaka https://www.stat.auckland.ac.nz/~ihaka/120/Lectures/lecture03.pdf

What is the story to tell?

• you don't care about the little plotting details
• plots are answering and raising questions

Which plot(s) tell the story best?

• different ways of mapping variables to aesthetics
• different subsets? summaries?

Match perceptual and data topology: e.g. shape is unordered, so it should be used for unordered variable

Make comparisons as easy as possible:

• Position on a common scale (Easiest)
• Position along identical, non-aligned scales
• Length
• Angle / Slope
• Area
• Volume
• Colour properties (Hardest)

How can I make it easy for the viewer?

• nice labeling,
• careful color choice,
• captions? annotations?
• version for screen or print?
• tweaks to axes, grid lines etc.

http://simplystatistics.org/2012/06/01/interview-with-amanda-cox-graphics-editor-at-the-new/