Measure 90 and the Equal Vote

Mark Frohnmayer of the Equal Vote Coalition speaks to a class at Lane Community College about ballot measure 90, the Oregon Top Two Open Primary.

[@b43xoit]

Error at 10:00. Some ranking systems, when they allow equal ranking, provide equality of voting weight. Not the Ware/Hare system, but some others.

[@UnifyTheVote]

Yes, this is correct. Ranking systems that allow equal ranks or require full rankings and also count those rankings equally comply with the equality criterion. This has been updated on the Equal Vote site with the description of Ranked Robin and others.

[@captsorghum]

Isn't this vote-reversal property the same as the symmetry property that Donald Saari found desirable about Borda voting? If so it seems to be at best one of many properties that could be considered desirable. I'm skeptical about any study that shows marked differences between approval- and score-based methods' Bayesian regret numbers, since optimal score strategy is the same as approval. Although even as shown here, there probably wasn't enough difference to bother about. Top-two approval and score voting methods are susceptible to adverse collusion strategy, where two smaller factions can use insincere voting strategies to knock a larger faction out of the runoff election, before ending their cooperation in the general. Better would be to keep partisan primaries and use approval voting in both the primary and general elections, so that each party primary sends one candidate to the general election. The primaries could be open so long as the voter can only participate in one primary at a time. Plain-vanilla approval voting is probably the only alternative to the current system that is simple enough to use in both primary and general elections.