On 20 Feb 2005 at 10:21 PST, Bart Ingles wrote: > Sorry, I didn't read the original proposal carefully enough to see that > it was intended as a type of non-partisan blanket primary. > > Open or blanket primaries make it easier to engage in pushover strategy, > where one party tries to make sure the opposing party nominates a weak > candidate who can be easily defeated in the general election. That's > the main reason most parties (e.g. Democrat, Republican, Libertarian) > oppose open primaries. A blanket primary was just defeated in > California (November 2004). > > Also, using an approval-like method in the first round of a two-round > election makes possible risk-free collusion strategies, where two > parties gang up to exclude a third (perhaps Condorcet candidate) so that > they can face each other in the final round. > > Would this proposal allow both strategies to be pursued simultaneously? > > My preference at present is for a plain approval primary, which is > either a closed primary or a California-style "modified open primary" in > which voters can choose any party on election day but can only > participate in that party's primary.
Yes, my original proposal was intended to be a replacement for a blanket primary. Washington State's blanket primary was removed by constitutional challenge (encouraged by Democrats then-Governor Locke and then-AG/current Governor Christine Gregoire). In principle, there is no reason to restrict the number of choices to 3, but when combined with approval (i.e., 1st/2nd/3rd vote == approval) it has the advantage of discouraging 'burying' strategies. However, I've been re-reading the issues surrounding the Washington State primary ballot, and have had to revise my thoughts a bit: - Major parties oppose the blanket primary. The current WA Secretary of State (Republican, but highly unpopular with his own party, might get recalled), favored the top-two runoff precisely because it discourages insincere cross-over voting. Since the WA version of top-two is different from the Louisiana version, it may face constitutional challenge. - Ballot initiatives can address only one issue at a time. So it isn't possible to change both the primary and general elections in one initiative. - The primary election (at least here) is sort of blind. At that point there is some information about the candidates but insufficient. But there have been many cases where a good compromise candidate has been eliminated by the primary. So here's an idea for modifying the proposal into something similar to top-two, but with a Condorcet-Approval twist: 1/2/3 ballot (equal ranking allowed), any 1/2/3 implies approval. At the top of the general election ballot, place the Smith Set, (ordered by Ranked Pairs); followed by any candidates with HIGHER approval than the minimum approval in the Smith Set, in descending order of approval; followed by the approval runner-up to the Smith Set. No automatic win. In many cases the Smith set would contain one member, the Condorcet winner, with highest approval. So there would be two candidates for the general election, the CW and the approval runner-up candidate. But there might still be cases in which the Condorcet winner doesn't have highest approval. So the general ballot would include the other candidates with higher approval, plus the approval runner-up to the CW. There could be 3 or more candidates on the final ballot. In the most pathological case, there might be a multi-candidate Smith set, one or more candidates with higher approval than the least-approved in the Smith set, and one more approval runner-up candidate. There could be 5 or more candidates on the final ballot. Would this still enable either pushover or risk-free collusion? If it enables either one, it would be opposed because it wouldn't meet the anti-crossover criterion. Would it be better to restrict this proposal to non-partisan races only? I think it would be best to have a Condorcet general election also, but that would require a separate proposal in order to withstand challenge. Ted -- Send real replies to ted stern at u dot washington dot edu Frango ut patefaciam -- I break so that I may reveal ---- Election-methods mailing list - see http://electorama.com/em for list info