Skip to:

Sparklines: Beautiful Evidence or Muddled Graphics?

Do you understand this graphic? It is an example of a sparkline, by Edward Tufte. Tufte was, if not the originator of sparklines, one of its earliest advocates. He wrote about them in his book Beautiful Evidence (2006); he defines sparklines as “small, intense, wordlike graphics, embedded in the context of words and numbers”. Tufte’s ideas were very influential and were taken up by Microsoft in their 2010 release of Excel. But I don't agree with him about sparklines. 

To be intelligible, graphics need space. In-line graphics, that is, graphics that appear on the same line as a line of text, are challenging in any form of publication, print or digital – the information that can be conveyed in such a compressed space is very limited, and the problem is only exacerbated with digital publishing.  My disagreement with Tufte, however, is that he appears to take a paradoxical delight in squeezing as much information as possible into this small space - I couldn’t understand his diagrams, and certainly not the diagram above.

The diagram is an example Tufte gives of sparklines to illustrate the results of a sports team. He uses a series of vertical lines to illustrate victories and defeats. A win is shown by a vertical above the centre of the line, while a defeat is shown by a vertical below it. The result is just about intelligible, no more than that. Yet, he claims, “Most sports fans will quickly derive the sparkline’s meaning from the context”, a statement that does not appear to have any evidence behind it. It is an indication of the rapid rise and acceptance of user experience tools that such statements would be unlikely today. 

I thought that fundamental principle of information graphics was to display one thing well. Yet Tufte recommends trying to convey multiple types of information in graphics: “a useful strategy for data displays is to multiply a good design” – in other words, try to show three things where you only have space to show one. Following his rule, Tufte attempts to cram more meaning into an already unclear sparkline: he adds colour to the sparklines to indicate the defeated team did not score; he adds horizontal lines to contrast home and away fixtures. All this within the leading of a text line! The result, frankly, is unintelligible. 

Microsoft Excel, example of sparkline

Paradoxically, the example given by Microsoft in a blog post for Excel 2010 take Tufte’s idea and make it clearer. They use blocks rather than lines. They use the colour red simply to indicate defeat. They reduce the complexity, in other words, and improve the intelligibility.  Here is proof that digital display can be clearer than a print equivalent, in the hands of a good designer.

In contrast, Tufte uses colour in other diagrams to indicate different conditions at the same time, for example the colour red used indicate both the oldest and the newest figures in a series (page 50).

So when Tufte concludes “Sparkline graphics give us some chance to be approximately right rather than exactly wrong” I disagree – sparkline graphics, by trying to cram multiple messages into a small space, have the potential to leave us confused rather than clear.