South Australian election morning after thread

The South Australian election result remains up in the air after a gravely disappointing night for the Liberals.

A grim night for the Liberals in South Australia, who needing six seats to win have only clearly won two from Labor (Bright and Hartley) and one from an independent (Mount Gambier), and will require sharp late-count reversals to add any more than Mitchell to that list. Labor can yet get to a majority if they can hold on to Mitchell, but the most likely result seems to be a hung parliament with the two returned independents, Geoff Brock in Frome and Bob Such in Fisher, deciding the issue. Another loser of the evening is “electoral fairness”, with the Liberals weak showing coming despite what looks like a win of about 52.5-47.5 on two-party preferred. I’ll have a lot more to say about this of course, but not right now.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

318 comments on “South Australian election morning after thread”

Comments Page 1 of 7
1 2 7
  1. Remember the Liberals were $1.01. Whatever happens from here it is another stick in the eye for the theory that election betting is all that predictive.

  2. OK, in Mitchell Labor is down by 148. They are not big electorates. Looks like a slightly higher %age of Lib vote in 2010 were postals cf ALP. So what is the evidence that Labor can hold this from here?

  3. There is absolutely no chance the ALP get to 24 seats.

    Bright, Hartley and Mitchell are gone, there is no coming back in any of these three seats.
    Adelaide has had a swing to Labor but it is not enough and declaration votes will favor Sanderson, no chance.

  4. Kevin, all the commentators I listened who commented on it stated that the declaration votes will favor the Liberals.

    This included Labor MP Michael Atkinson who said consistently throughout the night on the ABC Radio coverage that any seat with a Labor 2PP of less than 52 was not a certain win. That means seats like Colton, Newland and Elder are not absolutely certain wins although I think they will win all three.

    The maximum the Labor can get is 23, the probably will get it, but it is the maximum they can get.

  5. I hope Labor hold on in South Australia. The Libs are still not ready to govern this time around and next election should be a Labor wipeout.

  6. The one thing nobody has mentioned yet at this stage is the “informal votes”. These currently contain any ballot papers where the voter has not completely filled out the ballot paper. I know that in the Henley south booth at Colton there are quite a few papers where the voter chose “Paul Caica 1” and then didn’t bother filling out anymore of the paper. These will be reclassified as “formal” later on. I estimate this could add another 50 votes to labors overall tally on average per electorate.

    I also noticed this at the federal election, but under federal election rules, such ballot papers remain informal.

  7. Syksie: that is absolutely correct.

    I scrutineered in Florey and the “informal” first preference vote went to Bedford at a higher percentage than the “formal” first preference vote.

    Once these votes are counted back in, they will increase her lead.

    From my experience, the optional preferential voting system does tend to generally favour the incumbent; if it follows this time then that will help Labor in all those tight seats.

  8. PS your “average of 50 votes per electorate” estimate would be extremely conservative.

    The booth I was on was a shared booth and it was small … there were only 800 odd votes cast for Florey. Of these, Bedford got 13 “informal”. If you extrapolate that across the 14 booths (not to mention those that weren’t shared with another electorate and those that are very large)I expect the OP vote to add closer to 150-200 per electorate.

    Like I said, in tight seats, this could make a real difference.

  9. Effects of postal etc might be predicted by looking at the last federal senate election. For simplicity take the above the line votes for Labor/Liberal it was
    186,782 45.1%
    227,064 54.9%

    After adding in postal etc you get
    225,558 44.8%
    277,866 55.2%

    A .7% difference to the conservatives.

  10. The two party preferred is basically the same as Rudd’s win in 2007.

    How well do you think it would have gone down if Howie won that election with 47.5% of the vote?

    PB would be in riot.

  11. Scattered late-night thoughts:

    Who was it that reckoned the Powerful Communities guy was gonna do well in the upper house? Put your dunce cap on and sit in the corner, because he’s coming last. KAP are second last, even with a friend in the upper house. There goes Ann Bressington’s personal vote.

    The SA Nats also got splattered – 4.3% in Hammond (fifth and last), 1.8% in Goyder (sixth and last), 0.2% in the upper house (nineteenth and not quite last – hey, it worked for Wayne Dropulich 😛 ), didn’t even bother running in the two seats they’ve held in the past.

    The Greens came second in Heysen and get to have their first 2pp margin ever in SA, unless D4D voters heavily preferred Labor to the Greens for some reason.

    Meanwhile, the Greens got pwned hard in Giles – 12.1% down to 4.7% there. Craig Emerson could possibly torture a Whyalla-related joke out of that, although not a very good one.

    Dan van Longdutchname in Stuart got an 11.4% swing! That’s one helluva sophomore effect. If the Libs don’t win govt, he could get himself made leader for a little while (compare to Steven Marshall’s effort). I hope he does, just to harass political journos who already have to grapple with the Qld ALP leader’s name.

    D4D did OK for a microparty in the few seats they ran in – if they’d run in a few more, they might’ve gotten their upper house vote up enough for a seat there. Ah well.

    FF cracked double figures in seven seats:

    Napier 13.1
    Ramsay 11.4
    Reynell 10.7
    Taylor 10.6

    Chaffey 13.6
    Schubert 12.0
    Finniss 11.4

    They also had three candidate with the last name Hood. Family business, eh?

  12. Diogs,

    Happens all the time.

    Beazley lost in 1998 despite having a majority as did Peacock in 1990.

    Seems that the problem is that too many Libs only want to live in the vicinity of people like them.

  13. Diog –

    PB would be in riot.

    Yeah, it is a bit of a travesty.

    And the SA Electoral Commission is supposed to be obliged to redistribute to “fix” this … clearly an impossible task.

    PR is the way to go … MMP would be my choice.

  14. Martin Hamilton-Smith is calling for the system in SA to be changed and he used the gerrymander word, said it was almost like there was a gerrymander.

    He complained that the public have voted Liberal but they are not getting a Liberal government like the last election and 1989.

    I think he and other Liberals need to look at why they failed to win Ashford when it was there to be taken. They simply should not have had a swing against them like what happened in Ashford. Postal votes will make Ashford a bit closer but it is a huge failure by the Liberals that the seat is not clear win to them at this stage.

  15. GG

    It does happen a lot. Which doesn’t mean its not a bad thing. I think the system should reflect the will of the people and if one side clearly is favoured over the other, that side should win.

    It also affects the legitimacy of the government and how the electorate views the winners.


    Those three close ones are all 1.5% or more in Labors favour so apart from Mitchell, any more would be a long shot for the Libs.

  16. Spur212. I had a quick scan through the 2010 SA election report on ECSA’s website to see what happened in Mitchell. First, there’s no way to tell which votes excluded as informal, initially, get added back via the ticket vote provisions in the Act – if we accept Danny Lewis’s analysis, above, the 150 odd election night lead for the Liberal candidate would just about get wiped out. Interestingly, in 2010, Labor actually did slightly better on the declaration vote count than on ordinary votes to the tune of about half a percent (in Mitchell). Extrapolated to 2014, this would add around 25 2pp votes to the Labor total. So way way too close to call at this stage. Something else I noted more generally from the 2010 report is that in none of the marginals did Labor perform relatively poorly on declaration votes compared to ordinaries. With the closest election night margin being around 1.5% in any 2014 marginal currently included in Labor’s tally of 23 seats, that back of the envelope analysis would mean that none should be expected to shift through the declaration vote counting. I’m pretty comfortable that Labor has 23 with Mitchell being on a knife edge (if 2010 trends in declaration votes apply again). I was quite surprised to see how in 2010 there was so little difference between declaration vote and ordinary vote results across Labor’s marginals, which may reflect the strong ALP organisational structure in SA.

  17. Outsider

    You make some good points there. I only think they’ll win Mitchell because I live in the area and they did a coordinated campaign with Bright. I might be very wrong about it

  18. Dio

    The name was fine (you want a distinctive name in politics). The candidate however seemed a bit scary if you asked me

  19. An observation from touring a few booths in a few marginals: some of the Liberal people handing out how to vote cards seemed unusually attractive in a way that made me a bit suspicious. If it was one booth, I wouldn’t suspect a thing. Three booths in a row made me think they weren’t your average “grassroots” Liberal Party people

  20. the section of the SA Electoral Act which allows for a de facto ticket voting system in the lower house

    ick. How did that come about?

    The Federal Senate ticket voting I can understand. I think it has proven to be a big mistake but I can understand why it came about when they were looking for solutions to the high informal rate in Senate voting due to the complexity of voting for so many candidates.

    That can’t have been true, ever, for SA lower house voting.

  21. Bob Such will be speaker and based on current seat count, the ALP will retain power.

    On the winning the 2PP but losing the seat count, the problem is that the Libs have massive majorities in rural seats, that’s why and these people out in the sticks are very rusted on. How is the Libs 2PP in Adelaide metro where the majority of seats are?

  22. [What is wrong with the name Terina Monteagle? ]

    I’m sure Bertie Wooster was accidentally engaged to her at one stage.

  23. I didn’t think we had OPV in SA. I thought if you didn’t sequentially number all lower house boxes, it was informal.

  24. Dio it’s not really OPV. Just that if you vote, say, 1 for your preferred candidate and leave the other preferences blank, your vote gets counted as formal, provided your preferred candidate has lodged a ticket. The vote then gets counted in accordance with the preferences per the ticket. There is no potential preference exhaustion under this system as there is with OPV.

  25. Jackal the informal vote rate at the 2010 state election was 3.3% which is pretty low. The section 93 ticket vote system would account for this lower than expected informal vote rate. I wonder how many people just vote 1 in the expectation the vote will be informal, without knowing that the vote most likely will still be counted? From what I can work out it appears most lower house candidates lodge tickets.

  26. Outsider – that may well be the case, but having this dodgy bit of electoral arcana to reduce informal rates from 5-ish% to 3.3%? It’s just bizarre, and it seems very few people know about it. I’m not and never have been resident in SA so I have an excuse, but it seems politically engaged SAers here don’t know about it.

    Plus, if your intent is to allow these simple votes, why go with ticket voting rather than OPV? I’m guessing based on the Federal Senate ticket voting decision that OPV was either not considered in times gone by or was generally considered flawed or undesirable. With today’s perspective I don’t see how if the choice is between a ticket voting system and OPV that any sane person would choose the former as being “more democratic” or any other characteristic of a good voting system.

  27. Given easily the most likely result is for Libs to pick up Mitchell and nothing else, that makes Labor 23, Lib 22 Such and Brock.

    I can’t see how the Libs could form a stable government if that was the case. With one as Speaker, the floor would be 23-23, unless a Labor member became Speaker.

    I think Weatherill as Premier, Brock supporting Labor and Such as Speaker looks the best bet.

  28. My very dim recollection is that the ticket vote in the lower house has been around for a long time. Maybe 30 years or so? It’s not a big secret. All candidates and scrutineers should know about it. But I agree it’s not known about more widely nor is it publicised – hence my comment about what I was told when I was handed my ballot paper. I didn’t want to argue about it with the guy but thought it was curious!

  29. In 2010 there were 32,638 House votes admitted to the count by the ticket vote savings provision, or 3.3% of formal votes. The ticket vote savings provision has been used since 1985 and was introduced that year when ticket voting was first introduced for the Legislative Council. The experience of the 1984 Senate election had shown that Council ticket voting would induce ‘1’ only votes in the House, hence the provision. At every election since, the House ticket vote savings provision has halved the informal vote compared to if Federal formality rules had applied.

  30. Thanks Antony.

    Has there been a shift in thinking in terms of voting complexity from ticket voting to OPV? Was OPV considered for the Senate voting reform instead of ticket voting, and if so what was the thinking behind ticket voting being preferable?

  31. If the Liberals want the election result to be reflected proportionally in the Parliament they ought to support a change to PR in the lower house. After all, one could argue that neither Labor or the Liberals ought to receive a majority of the seats with a vote in the mid 30s or 40s. The current system highlights that the votes of those in marginal seats are ‘worth more’ than those in safe seats held by either party.

    The current gerrymander provisions requiring the redrawing of boundaries should be (but won’t be) abolished.

  32. Make of this what you will. In Mitchell, Hanna lodged a split ticket HTV, the Greens preference Labor and FF preference the Liberals.. These HTVs become very significant with the add back of the ticket votes.

  33. To someone from outside the state the two indies seem a parallel to Oakshot/Windsor (with one an obvious candidate for speaker).

    So it comes down to whether they believe their personal followings are stronger than the right wing leanings of their areas. ie are they an Oakshot who seemed to lose local support when he supported Labor or are they a Windsor who still stood a chance of winning if he had stood.

Comments Page 1 of 7
1 2 7

Leave a Reply

Your email address will not be published. Required fields are marked *