Blogging Senate forecasts and results in the WA Senate re-election until officially declared.

Twitter: @AU_Truth_Seeker


Wednesday 4 September 2013

Victorian Senate - TruthSeeker forecasts

I have updated my calculation of primary votes to allow for the latest polls, and slightly tweaked my method for calculating minor party support to allow for more national consistency.

See my previous post for the method I use within my model.

Most likely elected Senate (Victoria):

1 GRN, 2 ALP, 2 LNP, 1 FF

Primary votes:
LNP:       38.5% (+/-2.0%)
ALP:      33.9% (+/-1.8%)
GRN:      12.3% (+/-1.8%)
Minor parties:       15.3% comprising selected parties
FF:      2.3% (+/-1.1%)
WIKI:      0.9% (+/-0.45%)
Bullet Train:      0.09% (+/-0.045%) (see below for why I include this party)

Likelihood of election:
LNP:      2 elected, +17% chance of 3rd senator
ALP:      2 elected, +1% chance of 3rd senator
GRN:      99% likelihood
FF:      81%
Bullet Train:      2%
Bank Reform:      0.1%
Wikileaks:      0.00000%

Ideological splits:
3 Left, 3 Right: 98% likely
4* Left, 2 Right: 2% likely
*assuming the Bullet Train crew are more “Left” than “Right” - does anyone know?

Summary:
- Victoria appears set to elect a Family First senator… again! This can happen with FF vote being as low as 1%, depending on how other micro parties poll.
- Despite other pundits predicting an Assange Wikileaks victory, this remains highly unlikely even when the Wikileaks primary vote is as high as 1.35%.
- The scary prospect for Australians is the slim chance of “Bullet Train Australia” party.  According to my inputs, they are the 5th most likely party to be elected to the Senate from Victoria. While we don’t know that they’ll do a bad job, it is a big unknown.  I have commented previously about how democracy fails when any system allows for the election of any party with a minute small percentage of the vote. In Bullet Train’s case, as little as 0.1% or 4000 votes. This makes Family First’s estimated 1% seem massive, which it isn't.

As the state is the best in a 2PP sense for Labor, you would expect the guaranteed 3-3 split from Victoria, with a slim chance of 4-2 split is just about as good as it gets...

Over the coming days, I will publish equivalent summaries for other states. I won’t talk about the methodology – this is largely covered in prior posts. On election day, I will publish my full results, updated for latest polling data.




22 comments:

  1. What primaries are you giving Democratic Labour and Rise Up Australia? I would have thought that if they both poll at the upper end of their ranges there would be a small but non-zero chance of DLP taking the last "right" seat.

    ReplyDelete
    Replies
    1. Hi Nathan,
      DLP: 1.8% (range: 0.9% to 2.7%)
      RUA: 0.36% (range: 0.18% to 0.54%)
      I'm getting zero.

      Delete
    2. I'd probably expect higher than that... DLP is coming off a base of 2.35 and has an incumbent senator (Fielding increased the FF vote by about half a percent) and Rise up has lower house candidates in all Victorian seats IIRC. Of course the large number of minor parties complicates things. It would be nice to build a semi-reliable primary vote prediction model for minor parties, but as it is I suppose one guess is as good as another.

      Delete
    3. There's 34 candidates in VIC, 21 last time. The minor vote will scatter. Too much estimation to build a robust model

      Delete
  2. What are you using for Palmer and Katter?

    ReplyDelete
    Replies
    1. I think your PUP is on the low side.

      Delete
    2. Agree! 2% was reasonable in a very non-PUP state two days ago. When I publish re-estimates on election day, I will include a much higher PUP vote!
      I expect VIC to give PUP approx 50-70% of the national PUP vote, so perhaps 3-4% will be more reasonable.

      Delete
  3. Hello Truthseeker. I notice you use Bludgertrack numbers as a guide for your senate percentages. Have you done any analysis on the historical divergence between HOR and Senate party percentages? I'm wondering if there is a reduction factor you could apply to the major party vote.
    Also Senate ballot paper position for the micro-party in position A or the final position.

    ReplyDelete
    Replies
    1. Yes, analysed past results extensively. Check my earlier post "How to model.."
      I also apply a Group A donkey bonus of 0.3 to 0.5%, depending on the party. THis is in alignment with historical trends.

      Delete
  4. Would it be possible to run these numbers assuming that the Liberal Democrats, Smokers Rights, One Nation, etc had submitted tickets with similar preference flows to the other states (use NSW?)

    ReplyDelete
    Replies
    1. Great question. Yes, it would be possible, but it would take a fair bit of effort. I don't have the time for this at the moment unfortunately.

      Delete
  5. The analysis surrounding which, if any, minor party might take the sixth spot must be heavily dependent on your estimates for their relative primaries (and hence their order of exit in the early rounds of counting). I don't know of any robust methodology for estimating the performance of a new party? I do predict that you are wrong about Bullet train taking the sixth spot in the event of a 4L:2R split occurring.

    ReplyDelete
    Replies
    1. Bullet Train is the 4th left party, I arbitrarily assigned them as such because I assumed they'd be pro-train, pro-environment, anti-oil, etc! It's a big assumption, but minor. Should I reclassify them as left?

      Delete
    2. Sorry, should I reclassify them as right?

      Delete
    3. Turns out they're neither (except so far as bullet trains implicitly involve support for some things, and opposition to other things, as you mentioned). They deliberately have no policy at all on anything else: http://bullettrainforaustralia.com.au/faqs/

      At least they are a genuine party, i.e. can be arsed putting up a proper website and some HoR candidates and appear to be actually attached to their one issue, rather than being mere cynical vote-harvesters. I would still put their chances of winding up with a Victorian Senator at somewhat less than 2%, though!

      Delete
    4. Thanks for the research! I'll keep them as left as their single-issue is probably a left wing issue. What they think about asylum seekers is anyone's guess (Perhaps they'll support a fast train from Indonesia to PNG to speed processing? :-|

      Delete
  6. Maria Rigoni of the Bank Reform Party can win on several scenarios with 0.35%. It will be difficult for them to get as much as 0.35% in this election but it would be enough to win if they did.

    This is the one minor party that I would describe as not wacky. It is genuinely in the centre, which is why it has received a strong flow of preferences from both right and left.

    The Bank Reform Party was formed in WA by people who were customers of BankWest and affected by its collapse. They formed an action group to take up issues around regulation of the banking sector, and then registered a party to take these issues into the political arena.

    It has since expanded its agenda to include concentrations of market power in the retail sector, the insurance sector, the legal sector, as well as consumer issues in health, aged care and disability.

    They were about to change their name to something broader when the federal election was called. That will now be done after 7 September.

    ReplyDelete
    Replies
    1. Hi Vern.

      Thanks for your interest in my blog. I will correct you on a couple of points. Firstly, Bank Reform will not poll as high as 0.35%. Secondly, my Monte Carlo analysis shows that your candidate can be elected with just 0.14%. This is not something to be crowing about, it is instead a travesty of democratic justice. How is it democratic that a party with just a handful of votes can get elected? I think voters would be concerned about your plans to change your party's name...

      Some argue that for a party to get elected it must by definition get up to 14.3%. But this relies on our antiquated system of preferences whereby voters who think they are supporting something warm and fluffy instead have their vote hijacked and misappropriated to what may well be a party with diametrically opposed views.

      Democracy relies on the will of the people deciding who represents us. The current system does not support this principle.  

      Delete
  7. Both the lower house and the upper house have poor systems, and both need a good deal of renovation. We have a very poor system in the lower house where parties can win 20% across the country and not win a single seat. The system in the upper house also has its problems, in that a group like Family First can have a Senator Steve Fielding elected, for instance, with 1% of the vote, and voters subsequently discover a pentacostalist church has a member of parliament.

    The point is that BOTH the lower house system and the Senate system, as currently arranged, serve the Australian community poorly. Both need reform.

    The other key observation to make about your commentary is that minor parties are not necessarily wacky. A democratic system worth the name must allow new entrants. New entrants will inevitably begin small. Those that are mainstream and credible will grow, the others will remain small and on the fringe. The Bank Reform Party is not "my party", I am not a member of it. But its candidate is a more mainstream person than most of the major party hacks and apparatchiks.

    Having a Senate full of party officials and hacks is what makes it "unrepresentative swill". What is needed is representaton from real people with real jobs who do not live off taxpayer funded political machines. I hope the next Senate has less party hacks and more real people in it, whose occupations and social profiles more closely resemble the diversity of Australian society.

    ReplyDelete
    Replies
    1. Agree that reform is needed and that allowing parties free and fair entry needs to be integral to any reform. This needs to be balanced against the right of voters to be easily able to vote for the representatives of their choice.

      Apologies for associating you with the BRP. If I'd have paid more attention when I voted below the line I would have noticed you were the #2 candidate on the VOICE ticket in VIC. You'd be unsurprised to know my model has not had you as being elected in any of my simulations.

      Enjoy election day!

      Delete
  8. Good, I'm glad to hear this acknowledgement of the need for reform in both houses. I agree Voice has no chance of election or of development into a significant force and I'm fully aware of this.

    My interest is in development of a mainstream electoral alternative to the cartel that dominates our political culture. A lot of groundwork has been done over the past month towards an initiative in this direction after this election. It is a tragedy that the cartel has faced no mainstream, credible electoral alternative in this election but only a plethora of mostly personality-based and single-issue and fringe parties. This is like a world champion boxer getting into a ring and finding a baby wrapped in blue in the opposite corner.

    ReplyDelete