AV vs FPTP — the short(er) version « Gowers's Weblog

[Owen Densmore]

Brit friends of mine .. what is the impact of AV vs FPTP in your opinion?  Apparently the country voted against AV recently which seems a bit odd but then it may be difficult to communicate the advantage, thus folks stay with what they know.

My understanding is the FPTP is that the majority vote determines the winner, while AV is a multipass vote where you rank your preferences, the loser of each round is removed from the vote, and first to get to more than 50% wins.  It is one of the many "fair vote" ideas:

http://gowers.wordpress.com/2011/04/30/av-vs-fptp-the-shorter-version/

As I understand it, AV has the advantage of more evenly distributing the vote based on voter preference while FPTP tends to create heavy preference from slight preferences .. an "attractor".

        -- Owen

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

[Tom Carter]

Owen -

  Well, I'm not a Brit, but I have had an interest in voting (and even from a mathematician's perspective) . . .

  I was actually pretty disappointed in this whole process, and I'll have to say I found Gowers' posting(s) somewhat less than what they might have been.

  In Britain in recent years there actually has been active discussion of serious changes in their system for electing (the houses of) parliament.  There was real discussion of going to a "proportional representation" system:

     Suppose (nationally) there are 4 parties, with (statistical) support as follows:

                    A   30%
                    B   30%
                    C   25%
                    D   15%

      Assuming people actually voted this way, then under a proportional representation system, (the lower house of) parliament would be made up according to those percentages.  Implementation of this system would be something like:  each party would declare a prioritized list of "candidates."  After the election, the appropriate percentage of top "candidates" from each party, according to the party's share of the votes, would become members of parliament.  In practice, it would probably be up to each party to figure out on their own how to form their list.  In the election, a voter (probably) wouldn't vote for a "candidate," but rather for a "party" . . .

  Such a "proportional representation" system would (almost certainly) result in significant changes in parliament, and would (almost surely) be bad for parties A and B above.

  The two systems discussed in Gowers' post(s) really do nearly nothing to achieve "proportional representation" -- they're both basically "plurality vote" systems.  Under the so-called "AV" system (also sometimes called "instant runoff"), in the example above, in the first round D would be dropped (nearly everywhere . . .); in the second round, in most places C would be dropped, and A and B would divide the spoils, just like under the "FPTP" (current) systems.  As Gowers pointed out, the only real effect of a "choice" between AV and FPTP would be that (at least potentially, under AV) there would be less motivation (but not "no motivation") for people to vote "tactically" (e.g., don't vote for who you really want, because they have no chance anyway -- so vote instead for your second choice, who at least has a chance).  Unfortunately, part of what is going on here seems to be some confusion between "democracy" and "majoritarianism" (or some such).  We have seen over the years that "having open and free elections" (of whatever stripe) is not enough to "ensure actual democracy" (e.g., it would seem you also need free press, effective judicial system, effective financial regulation, relatively transparent government, etc., etc. . . .).  Declaring you have "democracy" because you have "majority rule" in your electoral process makes actual democracy a chimera at best . . . .

  What the process really boiled down to was a charade . . . have a "referendum on voting," but make sure that the only two (hmmm . . .) options don't actually address the problem, and either would leave the power structure in place.  After the referendum, the powers-that-be can declare that "the people have spoken" (whichever the outcome . . .), and so "there is no need to talk about this issue any more" (i.e., "shut up about this proportional representation stuff").

  Gowers seemed to have bought in to the false dichotomy of "FPTP" vs. "AV" and played the game of comparing the two, without any particular discussion of this being a largely bullsh*t choice.  He did acknowledge in one of his posts that:    
           Q. I see. So AV is a more proportional system.  

           A. Er, actually no. In fact, some have even suggested that it is less proportional.

but then went blithely along ignoring the issue . . .

  (Gowers was also pretty consistent about avoiding the "Arrow Voter Preference" theorem -- although it did show up in the comments on his blog posts a few times . . .)

  Oh, well . . .

tom

               
On Aug 27, 2011, at 9:48 AM, Owen Densmore wrote:

> Brit friends of mine .. what is the impact of AV vs FPTP in your opinion?  Apparently the country voted against AV recently which seems a bit odd but then it may be difficult to communicate the advantage, thus folks stay with what they know.
>
> My understanding is the FPTP is that the majority vote determines the winner, while AV is a multipass vote where you rank your preferences, the loser of each round is removed from the vote, and first to get to more than 50% wins.  It is one of the many "fair vote" ideas:
> http://gowers.wordpress.com/2011/04/30/av-vs-fptp-the-shorter-version/
>
> As I understand it, AV has the advantage of more evenly distributing the vote based on voter preference while FPTP tends to create heavy preference from slight preferences .. an "attractor".
>
>         -- Owen
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> lectures, archives, unsubscribe, maps at http://www.friam.org

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org