One of the lessons the 2000 election taught Americans is that voting machines can sometimes interfere with the conduct of elections. The Florida recount revealed that machine malfunctions (symbolized by hanging chad) and poor ballot design (symbolized by the butterfly ballot) can throw up barriers to voters who wish nothing more than to cast a ballot in an election.

The problems that beset Florida in 2000 were dramatic and compelling. Yet, once the controversy had died down, the serious question remained: Was what happened in Florida specific to the Sunshine State, or was it part of a more general malady that infected voting technology nationwide? Given the vastness of the country and the heterogeneity of how Americans vote, was there a way to put the problems brought to light in Florida in a broader context, so that this question about its general applicability to the rest of the nation could be answered?

The desire to place Florida’s problems counting votes into a broader national context led to the development of a metric to assess the relative performance of voting machines, termed the residual vote rate. This metric can be applied to any voting machine in such a way that it is possible to compare different machines with each other, as well as different communities. The purpose of this chapter is to explore this measure in some detail, so that its use can be better understood.