This graphic pretty much confirms my suspicions about the US presidential race as it stands at the moment.

From Real Clear Politics (via infographics news, I think it originally appeared in the Guardian but i can’t remember)

There are some problems - the spacing isn’t proportional but it pretends it might be, the lack of familiarity of the presentation (which can’t really be avoided) - but overall I think it’s a good graphic, boiling down a complex question into a simple glanceable chart. It’s obvious that if they want to win then the Republicans need to go with McCain.

The designers have been quite lucky with the data set but looking at the numbers reveals a couple of problems, for example in an Edwards-centric version of the chart Giuliani would appear above Huckabee, it’s easy to imagine a different set of figures that would throw up more such contradictions between figures and display so this is a decidedly un-portable approach.

Whether the visual lies undermine the chart I’m undecided, after all the data is all there in bold numerals.

[update: about 30 mins later]

So I spent a couple of minutes with a drawing package and came up with this based on the same figures…

Obviously it needs work, a description of what it does/ how to read it to start, but I think you should get the idea. It fixes some of the problems with the original but loses the nice leader board look. On the other hand it really does highlight that the ordering in the original was a bit of a sham. One of the other advantages of this formulation - Tufte would call it a small multiples presentation* - is that it’s portable, so you could run it from dynamic data.

The same graphical design structure is repeated for each of the twelve slices or multiples. Small multiples are economical: once viewers understand the design of one slice, they have immediate access to the data in all the other slices. Thus the eye moves from one slice to the next, the constancy of the design allows the viewer to focus on changes in graphical design.

[Visual Display of Quantitative Information]