You are pretending that there aren't circumstances that have an observed relation to systematic deviations from median election results that people can observe to understand that a wave election is more likely to happen in a particular election.