X marks the spot

There is many a psephologist who will tell you that polls of Senate voting intention are not to be taken seriously. However, a partial exception might be made for a poll of 542 Adelaide respondents, conducted by the Adelaide University politics department and published on the ABC site, in that it gives a unique gauge of support for Nick Xenophon’s Senate bid. The results are remarkable: support for Xenophon is at 24.2 per cent, even higher than his precedent-shattering 20.5 per cent at last February’s state election. While this would probably have been a little lower if the survey had also covered country areas, the poll provides evidence that Xenophon has his own seat in the bag, probably with enough of a surplus to deliver a seat to the Greens on preferences. The distant prospect of a seat for Xenophon’s running mate Roger Bryson has most likely been scotched by Labor’s decision to directly preference the Greens, although it’s still possible to construct scenarios in which the Greens’ seat goes to Bryson instead. The ABC report seems to suggest that Xenophon supporters will be forced to go below the line, which is not the case. Lower house voting intention is also covered, pointing to a swing to Labor of about 8 per cent.

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.

169 comments on “X marks the spot”

Comments Page 4 of 4
1 3 4
  1. Xenophon will still be elected, but because his name does not appear above the line – just Group S – and because of the Sunday Mail’s attacks on him, he may not have enough surplus to help his running mate get close. In the end this may help the Greens or the third ALP candidate.

  2. Frago61,

    I don’t think the DLP will get 1 per cent in SA. It’s more than 30 year since it last contested an election there, and even then SA was a weak state for it. It can poll only 2 per cent in Victoria, where it has had a continnuous presence since a minority of members continued its existence after the majority voted to close down in 1978 and where it now has an MLC who when both Labor parties vote together can defeat Opposition motions in the Legislative Council. I’d guess less then 0.5 per cent for it in SA.

  3. Phil,

    I agree, and I am still shocked at the Mail’s attack. It’s one thing to write a few negative stories on a candidate, another to spend an editorial having a crack at at a major (or even minor) party, but a completely different kettle of fish to spend an entire page taking, to use your words, an axe to an independent candidate. At least half of it was downright personal.

    Had it of said ‘we are disappointed in his decision, but he has made it now, there is no going back so good luck to him’ it would have been a different story, but that was disgraceful. What they are saying is that two years after endorsing him, they now think he should exit politics altogether?

    You’re right, wall to wall sleaze today – and I’m a liberal voter.


    A link for those who don’t live here

  4. I consider myself far left and give my support to around 95% of what Bill has to say.

    When discussing Mr X I think all you non-SAers must bear in mind how well known and popular he is here. His corflutes are very plentiful. He appeals to many because he is popularist. He is a weird mix of FF and the Greens, taking chunks from both of their agendas.

    There seems to be some confusion about Mr X’s split ticket, they go:
    Gr, FF, ALP, LIB


    FF, Gr, LIB, ALP

    So once FF get alliminated the entirety of his surplus go to the Greens.

  5. Dear Chris Curtis and Russ, thanks for your replies.

    Reducing the minor parties votes along the lines suggested (see below) gives

    1 FARRELL Don Australian Labor Party
    2 BERNARDI Cory Liberal Party
    3 XENOPHON Nick Independent
    4 WONG Penny Australian Labor Party
    5 BIRMINGHAM Simon Liberal Party
    6 HANSON-YOUNG Sarah Australian Greens

    with the first five being elected straight up (full quotas) and the Greens beating Family First (‘last standing but not seated’) into sixth place.

    Preferences ultimately flowing to Greens from SA, WWW, CCC, DEM, LDP, Grp B, SOL, ALP.

    Preferences ultimately flowing to Family First from AFLP, ON, ASP, NP, CDP, LIB, CEC, DLP.

    Preferences from Mr X (group s) evenly split.

    The minor party votes I put in can be regarded as ‘random values’ so a relatively
    large increase or decrease in any of their share of the vote, relative to each others, and the resultant change to the ultimate preference flows occurring could I suppose just as easily change the result (Family First instead of Greens) as a change to the primary results of those two parties themselves.

    All of which probably leads to the ‘brilliant’ conclusion that whatever the result it will be extremely close for sixth, and possibly fifth spot (if the Liberal vote is lower than expected).

    For me the benefit was really just to follow the various preference flows and thereby gain a better understanding of how the senate voting system works, courtesy of Mr Green’s calculator.


    Group A: One Nation .45
    Group B: Group B Independents .35
    Group C: Christian Democratic Party .36
    Group D: Australian Fishing and Lifestyle Party .44
    Group E: Australian Shooters Party .37
    Group F: The Greens 5
    Group G: National Party 1
    Group H: D.L.P. – Democratic Labor Party .43
    Group I: Liberal Party 30.5
    Group J: What Women Want .48
    Group K: Australian Labor Party 34.5
    Group L: Climate Change Coalition .42
    Group M: Citizens Electoral Council .5
    Group N: Senator On-Line .39
    Group O: Socialist Alliance .1
    Group P: Australian Democrats 1.9
    Group Q: Family First Party 3.4
    Group R: Liberty and Democracy Party .41
    Group S: Group S (Xenophon) 19

  6. Fargo61,

    I concur. The ABC Senate Calculator votes for SA do not equal 100 per cent, probably because groups which stood last time are not standing this time. I used these missing votes to fill up the groups, other than Group S, that did not stand in 2004 and then added a few more to the Liberals and the ALP in order to reach a total of 100 per cent. My purpose was to see what the different preference arrangements this time would do. The result was ALP 2, Liberals 3, Greens 1, with the Greens beating Family First for the last spot.

    I next put in a swing to the ALP along the lines of the Newspoll state figures. Newspoll shows a 9.6 per cent drop in the Liberal vote and a 16.5 per cent increase in the ALP vote. I settled on about a 10 per cent swing, putting the ALP on 46 per cent and the Liberals on 38 per cent. The result was ALP 3, Liberal 3, with the Liberals winning the last seat from the Greens on FF preferences.

    I next took account of the candidature of Nick Xenophon, using the figures from the SA state election as a rough guide. I gave him 18 per cent and cut the ALP by 8 per cent and the Liberals by 10 per cent. This is a comparatively low figure for Mr Xenophon. The result was ALP 2, Liberal 2, Mr Xenophon 1, Greens 1.

    I had to push the ALP vote up to 42 per cent and push other parties down for it to get a result of ALP 3, Liberal 2 and Mr Xenophon 1. The ALP is basically on its own in SA. It either polls very close to three quotas in its own right or it is stuck with two seats. Newspoll gives it 53 per cent, so it’s possible.


  7. I disagree that polls of Senate voting intention are not to be taken seriously, A well constructed/executed Senate poll is probably much more accurate then a nation wide two party preferred lower house poll. The results of the Senate should be known some time next week if Senate polling intentions are published.

    The ACT is looking very much like being a Coalition exclusion zone and yes SA will be interesting to watch as once Tasmania was.

    The Greens are still in a doubtful position in Victoria and the future will be determined by the extent that the Liberal party fall below the 42% threshold. if they receive 34% then the greens will be elected at the expense of a coalition no 3.

    The AEC afiled to deliver on their promiose of providing a csv file of registered above the preferences.

    If yiouy take time to look at the published pdf files you will noticce the inconsitenancy and porr data quaility mnagement.

    Missing still is statistics on the number of pre-polling votes issued and the number of postal votes that have been returned. (Hopefully they will be available next week)

    You can look at the postal votes issued data at http://www.aec.gov.au/Elections/federal_elections/2007/data_extracts/downloads/pva_stats_for_at071109.csv

  8. Max @ 34: Mr X has a Senate group ticket. He might not have running mates but he most certainly has a ticket.

    Analysis of the below the line votes shops that most voters remain within the group of their first choice an whilst they may diverge in final preference allocation in general they tend to follow more and less the same direction as the party ticket. Which is why it is reasonable to assume that a below the line vote is in effect party ticket. Those who vote number 1 for a lead candidate tend to get locked into the party flow.

    Those that vote one for the no 2 (Lib or ALP candidates) have their votes devalued as a result of the paper based surplus formula.

    So whilst the btl vote may register 5% you can expect that a much lower value then 5% actually changed the flow of preferences to the extent that it could bring about an upset outcome form analysis of the above the line tickets.

    In Victoria 2006 there was only one electorate that provided an change in the result as a result of below-the-line preferences that diverged from the party votes. And even then the result change was due to poor counting and votes that went missing between count A and count B and the adoption of optional preferential voting.

    I am of the view that if the results from count A and Count B is different then there should be automatically a count C to confirm that mistakes were not made in Count B. or as a minimum a distributed random sample of the count should be verified.

  9. Here is another unexpected ‘result’, (to me anyway) this time for my home state of QLD. I am posting it here in the absence of anywhere obviously more appropriate.

    Candidate Party
    1 MACDONALD Ian Douglas Liberal Party
    2 HOGG John Joseph Australian Labor Party
    3 BOYCE Sue Liberal Party
    4 MOORE Claire Australian Labor Party
    5 WATERS Larissa Australian Greens
    6 BUCHANAN Jeff Family First Party

    It comes from the following assumptions, the most ‘sensitive’ of which are that One Nation polls below Pauline Hanson, so that One Nation preferences flow to Family First, who then pick up ALP and Green (and various other) preferences to beat Ron Boswell (National Party) into sixth place. One Nation did poll below Mrs Hanson last time, so this seems reasonable.

    If Pauline Hanson polls below One nation, then her preferences go to One Nation and Family First get eliminated with Ron Boswell ultimately elected in sixth place, ahead of One Nation, on ALP and Green (and various other) preferences.

    I have also assumed an overall swing of about 4% away from Coalition to ALP.

    Party Total Votes %Votes Quotas
    Liberal/National Coalition 922559 39.24% 2.75
    Australian Labor Party 838156 35.65% 2.50
    The Greens 126957 5.40% 0.38
    Pauline’s United Australia 106738 4.54% 0.32
    Family First 79231 3.37% 0.24
    One Nation 73823 3.14% 0.22
    Australian Democrats 51723 2.20% 0.15
    The Fishing Party 30093 1.28% 0.09
    Senator On-Line 12695 0.54% 0.04
    Carers Alliance 12695 0.54% 0.04
    Climate Change Coalition 12460 0.53% 0.04
    Liberty and Democracy Party 12225 0.52% 0.04
    What Women Want 11990 0.51% 0.04
    Group K Independents 11520 0.49% 0.03
    Australian Shooters Party 11285 0.48% 0.03
    Group N Independents 11049 0.47% 0.03
    Australian Fishing & Lifestyle Party 10814 0.46% 0.03
    Non-Custodial Parents Party 4466 0.19% 0.01
    Citizens Electoral Council 3526 0.15% 0.01
    Group W Independents 2586 0.11% 0.01
    Socialist Alliance 2351 0.10% 0.01
    Group X Independents 2115 0.09% 0.01

  10. Fargo61:

    My thoughts on your numbers.

    Pauline’s vote will rise and One Nation’s drop. ONP had a senator up for re-election in 2004, and are on the far right of the ballot this time.

    Pauline: 5.5%, ONP: 1.5%.

    The fishing vote will be split between Lifestyle and plain Fishing. So both below 1%. Fishing will be higher due to left positioning and simpler name.

    Greens probably higher. Maximum 1%.

    I’d never even heard of Family First until half way through the 2004 campaign. They’ve had three years to build their profile.

    So Family First 1% higher.

    Your votes for some independent groups are quite generous.. Group W,K,N.. 0.3% maximum.

    Senator Online 0.1%.

    What Women Want have the prime left position. Might get 1%.

    Group X is James Baker, the National who contested pre-selection against Boswell. He’s had full page ads in country papers. I’m unsure what he’ll get but he won’t come last.

  11. Mr Speaker,

    I decided that I had best comply with your ruling (your logic seems pretty good to me) so I re-ran the figures (in Antony Green’s calculator) more along the lines that you suggested, and I also very slightly increased the Coalition vote. The end result is the same as previously! Family First beat Ron Boswell (National Party) into sixth place.

    It could be a very late night and a nerve wracking wait of a week or two, for some Senate candidates.

    1 MACDONALD Ian Douglas Liberal Party
    2 HOGG John Joseph Australian Labor Party
    3 BOYCE Sue Liberal Party
    4 MOORE Claire Australian Labor Party
    5 WATERS Larissa Australian Greens
    6 BUCHANAN Jeff Family First Party

    Party Total Votes % Votes Quotas
    Family First 356,115 15.15% 1.0602
    Liberal/National Coalition 315,612 13.42% 0.9396

    Party Total Votes % Votes Quotas
    Liberal/National Coalition 941,133 40.03% 2.8021
    Australian Labor Party 820,288 34.89% 2.4423
    The Greens 147,176 6.26% 0.4381
    Pauline’s United Australia 126,487 5.38% 0.3765
    Family First 100,625 4.28% 0.2995
    Australian Democrats 50,548 2.15% 0.1505
    One Nation 34,560 1.47% 0.1028
    The Fishing Party 23,040 0.98% 0.0685
    What Women Want 23,040 0.98% 0.0685
    Group X Independents 11,520 0.49% 0.0342
    Climate Change Coalition 9,404 0.40% 0.0279
    D.L.P. – Democratic Labor Party 9,169 0.39% 0.0272
    Carers Alliance 7,993 0.34% 0.0237
    Australian Fishing &Lifestyle Party 6,818 0.29% 0.0202
    Australian Shooters Party 6,818 0.29% 0.0202
    Liberty and Democracy Party 5,172 0.22% 0.0153
    Group W Independents 4,937 0.21% 0.0146
    Christian Democratic Party 4,702 0.20% 0.0139
    Group N Independents 4,701 0.20% 0.0139
    Non-Custodial Parents Party 4,466 0.19% 0.0132
    Citizens Electoral Council 3,526 0.15% 0.0104
    Group K Independents 2,586 0.11% 0.0076
    Socialist Alliance 2,351 0.10% 0.0069

  12. Much better Fargo.

    umm maybe a couple of points off the coalition vote into the labor column ?


    The key exclusion in the Queensland race is the Family First/Pauline one.

    The Greens -need- Family First to beat Pauline, otherwise her preferences go to Labor and push the Labor party ahead of the Greens and we end up with a 3-3 result.

    If FF beat Pauline they receive her preferences, which keeps the Greens ahead of Labor, and we end up with a 2-2-1-1.

    This also relies on the Coalition vote falling below 43%, which seems probable at this time.

  13. All the bleating about Pauline is simply that. In a boring campaign, something is needed to get voters to have a cursory look at the new Audi..

    Pauline is a spent force, that will nevertheless carry some weight for as long as she chooses to remain in politics until her rusted on supporters meet their maker. She was the face of One Nation, though even on 8% (entirely unrealistic) she fails to secure a seat.

    Much ridiculous press was made of FFPs “preferences with Pauline” in QLD, when they preferenced her 47 out of 65 candidates and the Coalition 7th position down!

    Now, Andrew Bartlett (DEMS) rattled the cage on this non-story and railed about her anti-muslim anti-immigration and racist policies. What he failed to be truthful about is that his same party in NSW preferenced Pauline ahead of Family First!!! This is despite Family First:

    1. Voting AGAINST the government on the terrible legislation to process refugees offshore (and Andrew was in the same room at the time!!!)
    2. Having policies to increase foreign aid
    3. Being PRO-IMMIGRATION and regardless of race nor religion!!

    Where are the supposed principles?

    In addition, both the Dems and Greens have made Climate Change central to this election. If they were truly being principled on this core issue, can they explain why they BOTH preferenced the ALP when they endorsed the pulp mill decision AND the Climate institute gave Family First a higher rating on their response to Climate change than either the ALP or the Libs???

    I just can’t believe this story got legs. It is really absurd.

  14. Unlike objective and restrained Greens, eh Bill? 🙂

    Spoken like a true conspiracy theorist. Are you on medication for the delusions at the moment?

Comments are closed.

Comments Page 4 of 4
1 3 4