Deepest Voting: a new way of electing

Classical approval voting processes suffer from well-known impossibility theorems. We consider grading-based voting, where voters give a grade rather than a preference to each candidate. Numerous voting processes have been proposed in this framework, such as majority judgment or range voting, showing interesting properties. This article aims to present a unified framework for grading-based voting processes. Our statement is that a voting process can be seen as the finding of the innermost point of a multivariate scatterplot. Introducing depth functions in voting theory enables to associate each voting process to a depth notion, and enlarges the scope of procedures. We focus on a given family of depth functions, the wL p depths, to highlight the behavior of the depth-based approach.