To all, Statistical tests are judged for their accuracy by how far they follow the scales of measurment (Sidney Siegal: Non-parametric statistics for the behavioral sciences). The four scales can also be applied to elections. (Later I found out that elections are statistical tests, that is in the sense that my innovation of Binomial STV is such). There is only one election system that follows all four scales, and that is transferable voting. Ranked choice or preference voting are indeed essential to an accurate election system: that covers the second scale: the ordinal scale. Proportional counting is also essential: that covers the fourth scale: the ratio scale. Strategic voting remains only a residual problem with STV. But it can occur in real life elections where a very popular candidate can take away most of the first preferences of an allied candidate, subjecting the ally to possible premature exclusion. Binomial STV solves that problem by making the exclusion count rational, as well as the election count. from Richard Lung. -- Richard Lung. http://www.voting.ukscientists.com Democracy Science series 3 free e-books in pdf: https://plus.google.com/106191200795605365085 E-books in epub format: https://www.smashwords.com/profile/view/democracyscience

On 12/20/2016 5:04 AM, Richard Lung wrote: > Strategic voting remains only a residual problem with STV. But it can > occur in real life elections where a very popular candidate can take > away most of the first preferences of an allied candidate, subjecting > the ally to possible premature exclusion. But surely (in normal STV) the surpluses are distributed before there are any exclusions, so wouldn't the surplus votes of the "very popular" candidate save the "allied" candidate from exclusion? Chris Benham On 12/20/2016 5:04 AM, Richard Lung wrote: > > To all, > > Statistical tests are judged for their accuracy by how far they follow > the scales of measurment (Sidney Siegal: Non-parametric statistics for > the behavioral sciences). The four scales can also be applied to > elections. (Later I found out that elections are statistical tests, > that is in the sense that my innovation of Binomial STV is such). > There is only one election system that follows all four scales, and > that is transferable voting. Ranked choice or preference voting are > indeed essential to an accurate election system: that covers the > second scale: the ordinal scale. Proportional counting is also > essential: that covers the fourth scale: the ratio scale. > > Strategic voting remains only a residual problem with STV. But it can > occur in real life elections where a very popular candidate can take > away most of the first preferences of an allied candidate, subjecting > the ally to possible premature exclusion. Binomial STV solves that > problem by making the exclusion count rational, as well as the > election count. > > from > Richard Lung. > > -- > Richard Lung. > http://www.voting.ukscientists.com > Democracy Science series 3 free e-books in pdf: > https://plus.google.com/106191200795605365085 > E-books in epub format: > https://www.smashwords.com/profile/view/democracyscience > > > > > ---- > Election-Methods mailing list - see http://electorama.com/em for list info > > > No virus found in this message. > Checked by AVG - www.avg.com <http://www.avg.com> > Version: 2016.0.7924 / Virus Database: 4739/13615 - Release Date: 12/19/16 >

On 12/20/2016 04:25 AM, C.Benham wrote: > On 12/20/2016 5:04 AM, Richard Lung wrote: > >> Strategic voting remains only a residual problem with STV. But it can >> occur in real life elections where a very popular candidate can take >> away most of the first preferences of an allied candidate, subjecting >> the ally to possible premature exclusion. > > But surely (in normal STV) the surpluses are distributed before there > are any exclusions, so wouldn't the surplus votes of the "very popular" > candidate save the "allied" candidate > from exclusion? It sounds like he's referring to Hylland free riding. Suppose candidate B is close to the quota and A is above the quota. Then voting A>B may fail to make B win, since after A gets elected, everybody who voted A first will have their voting power reduced. Thus voting B gives more power towards getting B elected if A is going to win anyway. The risk, of course, lies in overestimating A's support; and in that if everybody applies the strategy, the whole STV procedure may seem pointless because the final counts become pretty much the same as the initial counts.

You are right. I stand corrected. But there are safeguards against letting one of one's own get eliminated early. The key paragraph from the Electoral Reform Society pdf report on the Irish general election of 2016: "But campaigning in Ireland can often be even more local than this. When deploying multiple candidates in a constituency it is advantageous for a party to 'balance' their vote. This is done by strategically encouraging supporters to put different candidates as their 1st preference in different areas of a constituency, in order to make sure that candidates have relatively sizeable numbers of votes, so that none are eliminated early on by accident." This strategy can back-fire, presumably by over-estimating party support. Sinn Fein tried it early in the 2000s and let in one of their greatest opponents, a DUP candidate. On 20/12/2016 03:25, C.Benham wrote: > On 12/20/2016 5:04 AM, Richard Lung wrote: > >> Strategic voting remains only a residual problem with STV. But it can >> occur in real life elections where a very popular candidate can take >> away most of the first preferences of an allied candidate, subjecting >> the ally to possible premature exclusion. > > But surely (in normal STV) the surpluses are distributed before there > are any exclusions, so wouldn't the surplus votes of the "very > popular" candidate save the "allied" candidate > from exclusion? > > Chris Benham > > > On 12/20/2016 5:04 AM, Richard Lung wrote: >> >> To all, >> >> Statistical tests are judged for their accuracy by how far they >> follow the scales of measurment (Sidney Siegal: Non-parametric >> statistics for the behavioral sciences). The four scales can also be >> applied to elections. (Later I found out that elections are >> statistical tests, that is in the sense that my innovation of >> Binomial STV is such). There is only one election system that follows >> all four scales, and that is transferable voting. Ranked choice or >> preference voting are indeed essential to an accurate election >> system: that covers the second scale: the ordinal scale. Proportional >> counting is also essential: that covers the fourth scale: the ratio >> scale. >> >> Strategic voting remains only a residual problem with STV. But it can >> occur in real life elections where a very popular candidate can take >> away most of the first preferences of an allied candidate, subjecting >> the ally to possible premature exclusion. Binomial STV solves that >> problem by making the exclusion count rational, as well as the >> election count. >> >> from >> Richard Lung. >> >> -- >> Richard Lung. >> http://www.voting.ukscientists.com >> Democracy Science series 3 free e-books in pdf: >> https://plus.google.com/106191200795605365085 >> E-books in epub format: >> https://www.smashwords.com/profile/view/democracyscience >> >> >> >> >> ---- >> Election-Methods mailing list - seehttp://electorama.com/em for list info >> >> >> No virus found in this message. >> Checked by AVG - www.avg.com <http://www.avg.com> >> Version: 2016.0.7924 / Virus Database: 4739/13615 - Release Date: >> 12/19/16 >> > > > > ---- > Election-Methods mailing list - see http://electorama.com/em for list info -- Richard Lung. http://www.voting.ukscientists.com Democracy Science series 3 free e-books in pdf: https://plus.google.com/106191200795605365085 E-books in epub format: https://www.smashwords.com/profile/view/democracyscience