'Is AV really so complex? Or is it just confusion marketing?' blog discussion

edited 19 April 2011 at 10:04AM in Martin's Blogs & Appearances & MoneySavingExpert in the News
436 replies 25K views
1383940414244»

Replies

  • Sublime_2Sublime_2 Forumite
    15.7K Posts
    AV in principle is a good system. The trouble is when deciding which is the lesser of two evils with a party. Thats when it falls apart.

    I often have reservations about the party I vote for, so how much more confusing for all the other ones.
  • kermitfrogkermitfrog Forumite
    1.1K Posts
    Sublime wrote: »
    AV in principle is a good system.

    No it's not - it's a dog of a system. That's why it got such a drubbing.

    Now if you had had a referendum on PR, which has things going for it, we might have been looking at a dramatic change today.

    .
  • FATBALLZFATBALLZ Forumite
    5.1K Posts
    Confusion marketing only works on people who are easily confused. The sad fact is many people in this country are just not bright enough to understand AV, which is one of the arguments for keeping FPTP.
  • VT82VT82 Forumite
    1.1K Posts
    Tenth Anniversary 1,000 Posts Combo Breaker
    ✭✭✭
    Well the door is well and truly shut on AV. At least Martin and the other 'Vote yes' camp got their second choice...
  • MrChipsMrChips Forumite
    969 Posts
    Part of the Furniture 500 Posts Combo Breaker
    ✭✭✭
    mau408 wrote: »
    Exactly what I mean. Here's the example:

    If AV Decided Football Matches
    Round 1
    Each team plays each other
    Germany 1 1 Austria
    England 2 0 Scotland
    Scotland 0 0 Germany
    Austria 0 2 England
    Germany 3 0 England
    Austria 0 0 Scotland
    w d l f a pts
    England 2 0 1 4 3 6
    Germany 1 2 0 4 1 5
    Austria 0 2 1 1 3 2
    Scotland 0 2 1 0 2 2
    England do best over the whole campaign and win under first past the post.
    Scotland finish bottom and are eliminated.

    AV Round 2
    Scotland's Results are deleted
    Germany 1 1 Austria
    England 2 0 Scotland
    Scotland 0 0 Germany
    Austria 0 2 England
    Germany 3 0 England
    Austria 0 0 Scotland
    w d l f a pts
    Germany 1 1 0 4 1 4
    England 1 0 1 2 3 3
    Austria 0 1 1 1 3 1
    Scotland
    Germany Have Moved to first place under AV
    Austria Finish bottom and are eliminated

    If AV Decided Football Matches
    AV Round 3
    Austria's Results are deleted
    Germany 1 1 Austria
    England 2 0 Scotland
    Scotland 0 0 Germany
    Austria 0 2 England
    Germany 3 0 England
    Austria 0 0 Scotland
    w d l f a pts
    Germany 1 0 0 3 0 3
    England 0 0 1 0 3 0
    Austria
    Scotland

    Germany are the AV Winners
    5/6 Games did not matter in the end

    I like it how you assume I'm of the male variety kermitfrog!
    But laughed alot at amba_gamba although it seems a little harsh!

    My personal opinion is that some people are voting AV because people are calling them chicken for not wanting change. However I look at all sorts of sports tables and they are not really decided in this sort of way. I don't understand why a system such as political representation should be any different.

    Don't want to drag this on any more, but just want to clarify a misunderstanding here.

    The problem with this "example" is it is not first past the post either. First Past The Post works well in seats with only two candidates when it is a simple choice between them. The problem is when there are three or more it is not possible to establish the people's relative preferences between them.

    In some senses, the ideal voting system would be one where voters could identify preferences between each pair of opposing candidates and the relative preferences could be compared - this isn't practical in real life however, especially when many seats have upwards of 6 candidates. So in your example of 4 candidates, voters would be able to express a strong preference of Germany over England, of England over Scotland, neutral between Germany and Austria etc, and some sort of table could be put together...

    When asking people to name just one first choice (i.e. FPTP), it is impossible to ascertain if this is the “best fit” with the preferences of the electorate.

    In your example you assume England would win under FPTP as they top the table, but it could just as easily be Germany, or even in some situations (not based on your example) the team who comes bottom of the table.

    From your head to head results:
    We know people prefer Germany to England and are ambivalent between Germany and Austria and Germany and Scotland
    People prefer England to Scotland and Austria
    People are ambivalent between Scotland and Austria (a marginal preference for Austria on “goals scored”!)

    Under FPTP voters could come out 40% for Germany, 35% for England, and 15% for Austria and 10% for Scotland. Germany wins, even though when compared with other candidates, England are preferred on more occasions.

    Scotland voters would vote for anyone but England in a head to head (as usual!) but otherwise tend to back the stronger team, England voters would tend to vote for anyone but Germany but otherwise tend to back the underdog. German and Austrian voters are more balanced and usually (but not exclusively) vote based on which team they feel is stronger.

    These voting patterns would give the head to head results as shown in your example, which suggests that England is the deserving winner (although as this example suggests it depends on how you measure it), but with Germany winning under FPTP.

    Interestingly, to my mind AV is more like your “head to head” example as it whittles down the candidates until there is a face off between just two with the electorate given the opportunity to make their choice between the two most popular/least hated.
    If I had a pound for every time I didn't play the lottery...
  • MrChipsMrChips Forumite
    969 Posts
    Part of the Furniture 500 Posts Combo Breaker
    ✭✭✭
    As an example of a case where the team bottom of the group win under FPTP...

    4 teams/candidates, called A, B, C and D

    60 % of electorate prefer B to A i.e. a score of, say, A 0 - 2 B
    60 % of electorate prefer C to A i.e. A 1 - 3 C
    60 % of electorate prefer D to A i.e. A 2 - 4 D
    Electorate are split equally on the merits of B, C and D, a 1-1 draw in all games between them.

    Final table
    D won 1, drew 2, lost 0, points 5 (top on goals scored)
    C won 1, drew 2, lost 0, points 5
    B won 1, drew 2, lost 0, points 5
    A won 0, drew 0, lost 3, points 0

    Candidate D scores a marginal victory here. Candidate A is roundly thrashed.

    Under FPTP where only first choices are measured, candidate A is favoured by 40% of electorate, but the remaining 60% are united in their digust of A and are split evenly (i.e. 20% each) for candidates B, C and D. Candidate A wins.

    This may seem an example which is unlikely in practice, but can arise in situations such as the expenses scandal where an MP in a safe seat can rely on a sizeable vote from supporters of his party, no matter what his behaviour, while the other voters vote for "anyone else but him/her" and have their votes split amongst the various alternatives.
    If I had a pound for every time I didn't play the lottery...
This discussion has been closed.
Latest News and Guides

£10-£50 London theatre tickets

For over 45 shows in January and February

MSE Deal

Join the 1p Savings Challenge 2022!

It should help you save £667.95

MSE Forum