In XA (aka ATDT) you should vote your Top set at Top, and your Bottom set Bottom. But there is less incentive to vote the remaining candidates near the extremes as compared to MJ/Bucklin. > > From: Michael Ossipoff <email9648742@gmail.com> > To: EM <election-methods@lists.electorama.com> > Subject: [EM] Forest: ER Bucklin vs ATDT > Message-ID: > <CAOKDY5AOPc1939rf-1sGSsfWMSeLc=5rC8qvL8RwthWkxmZ > 5WA@mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Forest-- > > You mathematicians are more versatile than most of us, and can perceive > things that for many, are way too confusing. The mathematicians' > descriptions & discussions of XA are confusing for me, which is why I just > call that method "Approve To Desired Total" (ATDT). > > The version of ER Bucklin that is now my favorite, with equal-ranking > anywhere allowed, and with skipping allowed, and without that automatic > skipping that I used to suggest (for preserving MMC) is, of course > equivalent to MJ. > > You said that ATDT is better than ER Bucklin/MJ because, with an electorate > polarized about a candidate, ER Bucklin tends to put that candidate at top > or bottom, rather than at a more middle place that better reflects hir > average likedness. > > But I _want_ to send the good candidates to top, and the bad candidates to > bottom. > > So how is ATDT better than ER Bucklin, as regards voting strategy? > > Michael Ossipoff >

Yes, the inbetween candidates are the problem, when there are some who are almost tops, but they have a flaw, and, due to rivalry, their voters are likely not vote for your most-preferred candidates. In the ER Bucklin that I like: As you said, I'd top-rate the strong top-set & bottom-rate the strong bottom-set. As for the inbetweens, the tops-with-a-flaw and untrustworthy voters, I'd, for each of them, skip a few levels, and rank them below. I'd try to skip just enough levels so that the candidates better than them would have fully enough rounds to receive consolidated support from preferers of other candidates, before a vote is given to the inbetween candidate. The skipping-distances would be different for the different inbetween candidates, depending on their undeservingness & their voters untrustworthyness. ER Bucklin should have unlilmited rating-levels available. Maybe, instead of actually showing each level on the ballot, the voter could indicate that s/he wants to skip a certain number of levels between hir ratings of any 2 candidates. Michael Ossipoff On Mon, Nov 7, 2016 at 5:03 PM, Forest Simmons <fsimmons@pcc.edu> wrote: > In XA (aka ATDT) you should vote your Top set at Top, and your Bottom set > Bottom. But there is less incentive to vote the remaining candidates near > the extremes as compared to MJ/Bucklin. > >> >> From: Michael Ossipoff <email9648742@gmail.com> >> To: EM <election-methods@lists.electorama.com> >> Subject: [EM] Forest: ER Bucklin vs ATDT >> Message-ID: >> <CAOKDY5AOPc1939rf-1sGSsfWMSeLc=5rC8qvL8RwthWkxmZ5WA@mail. >> gmail.com> >> Content-Type: text/plain; charset="utf-8" >> >> Forest-- >> >> You mathematicians are more versatile than most of us, and can perceive >> things that for many, are way too confusing. The mathematicians' >> descriptions & discussions of XA are confusing for me, which is why I just >> call that method "Approve To Desired Total" (ATDT). >> >> The version of ER Bucklin that is now my favorite, with equal-ranking >> anywhere allowed, and with skipping allowed, and without that automatic >> skipping that I used to suggest (for preserving MMC) is, of course >> equivalent to MJ. >> >> You said that ATDT is better than ER Bucklin/MJ because, with an >> electorate >> polarized about a candidate, ER Bucklin tends to put that candidate at top >> or bottom, rather than at a more middle place that better reflects hir >> average likedness. >> >> But I _want_ to send the good candidates to top, and the bad candidates to >> bottom. >> >> So how is ATDT better than ER Bucklin, as regards voting strategy? >> >> Michael Ossipoff >>

As for voting in MDDTR, of course, as always, one would top-rate one's strong top-set, and bottom-rate one's strong bottom-set. For the inbetweens, if there are any (and there probably would be some, for me): Top-Count: Whan you rank someone below top, you don't give them a top-count vote, for the comparison of the un-disqualified candidates. But, ideally, if there's no indifference, then there'd only be one un-disqualified candidate: The CWs. Anyway, a demoted candidate can't be expected to be fully supported as well as a strong top-set candidate. A top-vote could matter, too, if everyone were majority-beaten. But that's a relatively unlikely snafu, especially if the 1st-choice supporters of the CWse (expected or evident CWs) are defensively plumping, or if there's even a convincing threat that they will.. Majority Disqualifications: II'd probably rank all of the inbetweens sincerely, below top & above bottom. ...confident that, if there isn't a significant amount of indifference about the CWs (and surely that's the least likely candidate for people to be indifferent about), and if there isn't successful pairwise-count offensive strategy (Truncation won;t work, and burial is deterred by the threat of defensive plumping), then the CWs will win. ...just by virtue of the majority-disqualifications. Voting a candidate over those less-liked, thereby contributing to the majority-disqualification of those less-liked, is all the protection that you need to give that candidate. Yes MDDTR needs to not have significant indifference about the CWs, for it to work well, but, for an electorate not indifferent about the CWs, MDDTR avoids the big avoidable problems, by meeting FBC, CD, & having wv-like strategy. Michael Ossipoff On Mon, Nov 7, 2016 at 5:46 PM, Michael Ossipoff <email9648742@gmail.com> wrote: > Yes, the inbetween candidates are the problem, when there are some who are > almost tops, but they have a flaw, and, due to rivalry, their voters are > likely not vote for your most-preferred candidates. > > In the ER Bucklin that I like: > > As you said, I'd top-rate the strong top-set & bottom-rate the strong > bottom-set. > > As for the inbetweens, the tops-with-a-flaw and untrustworthy voters, I'd, > for each of them, skip a few levels, and rank them below. I'd try to skip > just enough levels so that the candidates better than them would have fully > enough rounds to receive consolidated support from preferers of other > candidates, before a vote is given to the inbetween candidate. The > skipping-distances would be different for the different inbetween > candidates, depending on their undeservingness & their voters > untrustworthyness. > > ER Bucklin should have unlilmited rating-levels available. Maybe, instead > of actually showing each level on the ballot, the voter could indicate that > s/he wants to skip a certain number of levels between hir ratings of any 2 > candidates. > > Michael Ossipoff > > > > On Mon, Nov 7, 2016 at 5:03 PM, Forest Simmons <fsimmons@pcc.edu> wrote: > >> In XA (aka ATDT) you should vote your Top set at Top, and your Bottom set >> Bottom. But there is less incentive to vote the remaining candidates near >> the extremes as compared to MJ/Bucklin. >> >>> >>> From: Michael Ossipoff <email9648742@gmail.com> >>> To: EM <election-methods@lists.electorama.com> >>> Subject: [EM] Forest: ER Bucklin vs ATDT >>> Message-ID: >>> <CAOKDY5AOPc1939rf-1sGSsfWMSeLc=5rC8qvL8RwthWkxmZ5WA@mail.gm >>> ail.com> >>> Content-Type: text/plain; charset="utf-8" >>> >>> Forest-- >>> >>> You mathematicians are more versatile than most of us, and can perceive >>> things that for many, are way too confusing. The mathematicians' >>> descriptions & discussions of XA are confusing for me, which is why I >>> just >>> call that method "Approve To Desired Total" (ATDT). >>> >>> The version of ER Bucklin that is now my favorite, with equal-ranking >>> anywhere allowed, and with skipping allowed, and without that automatic >>> skipping that I used to suggest (for preserving MMC) is, of course >>> equivalent to MJ. >>> >>> You said that ATDT is better than ER Bucklin/MJ because, with an >>> electorate >>> polarized about a candidate, ER Bucklin tends to put that candidate at >>> top >>> or bottom, rather than at a more middle place that better reflects hir >>> average likedness. >>> >>> But I _want_ to send the good candidates to top, and the bad candidates >>> to >>> bottom. >>> >>> So how is ATDT better than ER Bucklin, as regards voting strategy? >>> >>> Michael Ossipoff >>>