The change from essentially two party contests to three-, or even four-cornered fights, has provided the impetus for a fresh look at both diagrammatic representations of an election outcome, and the measure of swing from one election to the next. The representation of the votes in a three-cornered contest has been considered, amongst others, by Ibbetson [1], by Hollingdale [2] and by Miller in a series of papers and books (see for example [3] and [4]). Miller puts forward three ways of describing the outcome of such a contest, the ‘box-corner’ method (three-dimensional coordinate axes), homogeneous coordinates, and the ‘rectangular coordinate derivation’, but, as he notes, these three are fundamentally the same. Miller states that ‘mathematicians will instinctively go for the homogeneous coordinate representation’, and Hollingdale kindly obliges, although Miller himself plumps for the third representation, [3].