Hi Forest (and everyone else on EM), As I found out today, the Texas Green Party is using PAV for delegate counts. Someone from the party joined the Discord server operated by The Center for Election Science, and is trying to figure out vote totals for PAV. Remind me to send a summary later if I don't follow up. I'm still in a real-time discussion as of right now. Here's a direct link to get an invite to the Discord server (if you aren't on it) https://discord.gg/vxzeh4B Rob

Thanks. I'm very glad to see that! I'm a huge fan of PAV. I joined the Discord discussion, which notes: > The Green Party of Texas are supposed to use "Proportional Approval Voting" for purposes of delegate apportionment for allocating our 26 delegates for the Presidential nomination. The Green Party candidates for president are (first names): Howie Dario Kent Sedinam Susan Dennis David Chad and NOTA. See details at https://en.wikipedia.org/wiki/2020_Green_Party_presidential_primaries A copy of the ballots etc. was shared on Discord. There I noted that one way to do this sort of thing with PAV would be to have delegates explicitly run. Then people can approve of however many delegates they want, and you pick the top 26 via PAV. Then the delegates can use whatever scheme you want to select the overall winner. But that would be an unusual procedure. Note also that I have put Python code for PAV up at https://github.com/nealmcb/pr_voting_methods There I also analyzed some real-world data from Block Plurality elections in Colorado via PAV. But back to the delegate allocation question at hand... I dare say the question conventionally is more like an apportionment procedure, as used for allocating seats in congress to states, or seats in the legislatures of most countries to parties. In many countries, they use "highest averages methods" (also known as "divisor methods"). Democrats in the US use the largest remainder method (also known as Hare–Niemeyer method, Hamilton method or as Vinton's method). Both have strengths and weaknesses under different circumstances: https://en.wikipedia.org/wiki/Highest_averages_method https://en.wikipedia.org/wiki/Largest_remainder_method Recently I ran across what seems to be a grand-unification theory of sorts, I guess. Multiwinner Approval Rules as Apportionment Methods Markus Brill University of Oxford mbrill@cs.ox.ac.uk Jean-Franc¸ois Laslier Paris School of Economics jean-francois.laslier@ens.fr Piotr Skowron University of Oxford piotr.skowron@cs.ox.ac.uk https://arxiv.org/pdf/1611.08691.pdf > We establish a link between multiwinner elections and apportionment problems by showing how approval-based multiwinner election rules can be interpreted as methods of apportionment. We consider several multiwinner rules and observe that they induce apportionment methods that are well-established in the literature on proportional representation. For instance, we show that Proportional Approval Voting induces the D’Hondt method and that Monroe’s rule induces the largest remainder method. We also consider properties of apportionment methods and exhibit multiwinner rules that induce apportionment methods satisfying these properties. But I haven't digested it yet. Thoughts? Neal McBurnett http://neal.mcburnett.org/ On Sat, Mar 14, 2020 at 8:54 PM Rob Lanphier <robla@robla.net> wrote: > Hi Forest (and everyone else on EM), > > As I found out today, the Texas Green Party is using PAV for delegate > counts. > Someone from the party joined the Discord server operated by The > Center for Election Science, and is trying to figure out vote totals > for PAV. > > Remind me to send a summary later if I don't follow up. I'm still in > a real-time discussion as of right now. > > Here's a direct link to get an invite to the Discord server (if you > aren't on it) > https://discord.gg/vxzeh4B > > Rob > ---- > Election-Methods mailing list - see https://electorama.com/em for list > info > -- Neal McBurnett http://neal.mcburnett.org/

I think Example 7 in that paper ("Sequential PAV as an apportionment method") closely maps to the Texas situation. I suspect you essentially come up with a large set of induced candidates via 26 "parties" (one per delegate), where each party has 9 candidates (one for each of their presidential candidates, plus NOTA). That's 234 "induced candidates". And you can induce votes for those candidates, run the PAV algorithm on it, and figure out an equivalent apportionment as described in the paper. But for now I need to get back to other things.... On Sun, Mar 15, 2020 at 9:53 AM Neal McBurnett <nealmcb@gmail.com> wrote: > Thanks. I'm very glad to see that! I'm a huge fan of PAV. > > I joined the Discord discussion, which notes: > > > The Green Party of Texas are supposed to use "Proportional Approval > Voting" for purposes of delegate apportionment for allocating our 26 > delegates for the Presidential nomination. > > The Green Party candidates for president are (first names): Howie Dario > Kent Sedinam Susan Dennis David Chad and NOTA. > > See details at > https://en.wikipedia.org/wiki/2020_Green_Party_presidential_primaries > > A copy of the ballots etc. was shared on Discord. > > There I noted that one way to do this sort of thing with PAV would be to > have delegates explicitly run. Then people can approve of however many > delegates they want, and you pick the top 26 via PAV. Then the delegates > can use whatever scheme you want to select the overall winner. But that > would be an unusual procedure. > > Note also that I have put Python code for PAV up at > https://github.com/nealmcb/pr_voting_methods > > There I also analyzed some real-world data from Block Plurality elections > in Colorado via PAV. > > But back to the delegate allocation question at hand... > > I dare say the question conventionally is more like an apportionment > procedure, as used for allocating seats in congress to states, or seats in > the legislatures of most countries to parties. > > In many countries, they use "highest averages methods" (also known as > "divisor methods"). > Democrats in the US use the largest remainder method (also known as > Hare–Niemeyer method, Hamilton method or as Vinton's method). > > Both have strengths and weaknesses under different circumstances: > > https://en.wikipedia.org/wiki/Highest_averages_method > https://en.wikipedia.org/wiki/Largest_remainder_method > > Recently I ran across what seems to be a grand-unification theory of > sorts, I guess. > > Multiwinner Approval Rules as Apportionment Methods > Markus Brill University of Oxford mbrill@cs.ox.ac.uk Jean-Franc¸ois > Laslier Paris School of Economics jean-francois.laslier@ens.fr Piotr > Skowron University of Oxford piotr.skowron@cs.ox.ac.uk > https://arxiv.org/pdf/1611.08691.pdf > > > We establish a link between multiwinner elections and apportionment > problems by showing how approval-based multiwinner election rules can be > interpreted as methods of apportionment. We consider several multiwinner > rules and observe that they induce apportionment methods that are > well-established in the literature on proportional representation. For > instance, we show that Proportional Approval Voting induces the D’Hondt > method and that Monroe’s rule induces the largest remainder method. We also > consider properties of apportionment methods and exhibit multiwinner rules > that induce apportionment methods satisfying these properties. > > But I haven't digested it yet. > > Thoughts? > > Neal McBurnett http://neal.mcburnett.org/ > > On Sat, Mar 14, 2020 at 8:54 PM Rob Lanphier <robla@robla.net> wrote: > >> Hi Forest (and everyone else on EM), >> >> As I found out today, the Texas Green Party is using PAV for delegate >> counts. >> Someone from the party joined the Discord server operated by The >> Center for Election Science, and is trying to figure out vote totals >> for PAV. >> >> Remind me to send a summary later if I don't follow up. I'm still in >> a real-time discussion as of right now. >> >> Here's a direct link to get an invite to the Discord server (if you >> aren't on it) >> https://discord.gg/vxzeh4B >> >> Rob >> ---- >> Election-Methods mailing list - see https://electorama.com/em for list >> info >> > > > -- > Neal McBurnett http://neal.mcburnett.org/ > -- Neal McBurnett http://neal.mcburnett.org/

Hi Neal Thank you so much for jumping in! I haven't quite figured out how to make PAV work for the election as described on the Texas Greens' website: http://www.txgreens.org/bylaws My fear is that they painted themselves in a corner by mandating the use of PAV without fully understanding how to apply the system. I suspect a couple of things: 1. Your proposal is their best path forward 2. The folks who wrote the bylaws hadn't anticipated your solution It'll be interesting to see how this plays out. I'm also probably going to spend some time on other things as well, but I'm interested to see how this plays out. It seems likely that things can go a lot more smoothly for them in 2024 if they spend some time on the bylaws between now and then. Rob Rob On Sun, Mar 15, 2020 at 11:48 AM Neal McBurnett <nealmcb@gmail.com> wrote: > > I think Example 7 in that paper ("Sequential PAV as an apportionment method") closely maps to the Texas situation. > > I suspect you essentially come up with a large set of induced candidates via 26 "parties" (one per delegate), where each party has 9 candidates (one for each of their presidential candidates, plus NOTA). That's 234 "induced candidates". And you can induce votes for those candidates, run the PAV algorithm on it, and figure out an equivalent apportionment as described in the paper. > > But for now I need to get back to other things.... > > On Sun, Mar 15, 2020 at 9:53 AM Neal McBurnett <nealmcb@gmail.com> wrote: >> >> Thanks. I'm very glad to see that! I'm a huge fan of PAV. >> >> I joined the Discord discussion, which notes: >> >> > The Green Party of Texas are supposed to use "Proportional Approval Voting" for purposes of delegate apportionment for allocating our 26 delegates for the Presidential nomination. >> >> The Green Party candidates for president are (first names): Howie Dario Kent Sedinam Susan Dennis David Chad and NOTA. >> >> See details at https://en.wikipedia.org/wiki/2020_Green_Party_presidential_primaries >> >> A copy of the ballots etc. was shared on Discord. >> >> There I noted that one way to do this sort of thing with PAV would be to have delegates explicitly run. Then people can approve of however many delegates they want, and you pick the top 26 via PAV. Then the delegates can use whatever scheme you want to select the overall winner. But that would be an unusual procedure. >> >> Note also that I have put Python code for PAV up at https://github.com/nealmcb/pr_voting_methods >> >> There I also analyzed some real-world data from Block Plurality elections in Colorado via PAV. >> >> But back to the delegate allocation question at hand... >> >> I dare say the question conventionally is more like an apportionment procedure, as used for allocating seats in congress to states, or seats in the legislatures of most countries to parties. >> >> In many countries, they use "highest averages methods" (also known as "divisor methods"). >> Democrats in the US use the largest remainder method (also known as Hare–Niemeyer method, Hamilton method or as Vinton's method). >> >> Both have strengths and weaknesses under different circumstances: >> >> https://en.wikipedia.org/wiki/Highest_averages_method >> https://en.wikipedia.org/wiki/Largest_remainder_method >> >> Recently I ran across what seems to be a grand-unification theory of sorts, I guess. >> >> Multiwinner Approval Rules as Apportionment Methods >> Markus Brill University of Oxford mbrill@cs.ox.ac.uk Jean-Franc¸ois Laslier Paris School of Economics jean-francois.laslier@ens.fr Piotr Skowron University of Oxford piotr.skowron@cs.ox.ac.uk >> https://arxiv.org/pdf/1611.08691.pdf >> >> > We establish a link between multiwinner elections and apportionment problems by showing how approval-based multiwinner election rules can be interpreted as methods of apportionment. We consider several multiwinner rules and observe that they induce apportionment methods that are well-established in the literature on proportional representation. For instance, we show that Proportional Approval Voting induces the D’Hondt method and that Monroe’s rule induces the largest remainder method. We also consider properties of apportionment methods and exhibit multiwinner rules that induce apportionment methods satisfying these properties. >> >> But I haven't digested it yet. >> >> Thoughts? >> >> Neal McBurnett http://neal.mcburnett.org/ >> >> On Sat, Mar 14, 2020 at 8:54 PM Rob Lanphier <robla@robla.net> wrote: >>> >>> Hi Forest (and everyone else on EM), >>> >>> As I found out today, the Texas Green Party is using PAV for delegate counts. >>> Someone from the party joined the Discord server operated by The >>> Center for Election Science, and is trying to figure out vote totals >>> for PAV. >>> >>> Remind me to send a summary later if I don't follow up. I'm still in >>> a real-time discussion as of right now. >>> >>> Here's a direct link to get an invite to the Discord server (if you >>> aren't on it) >>> https://discord.gg/vxzeh4B >>> >>> Rob >>> ---- >>> Election-Methods mailing list - see https://electorama.com/em for list info >> >> >> >> -- >> Neal McBurnett http://neal.mcburnett.org/ > > > > -- > Neal McBurnett http://neal.mcburnett.org/