Photo finishes: Denison

Saturday, August 28

The AEC has published its provisional distribution of preferences which makes it very clear that Andrew Wilkie will surpass the Liberal candidate in very fine style, recording 20,430 votes to the Liberals’ 15,695 after distribution of Greens and Socialist Alliance preferences, and then comfortably winning the seat on Liberal preferences.

Friday, August 27

The Australian Electoral Commission announces it will conduct a “provisional” distribution of preferences in Denison to ascertain whether the Liberals are likely to be excluded from the count before Andrew Wilkie, a necessary precondition for the latter winning the seat.

Tuesday, August 24

6pm. Indicative preference count finished for real now, with pre-polls and hospital booths added, and Wilkie’s lead has risen to 1.2 per cent (1375 votes).

3pm. The indicative preference count for ordinary votes has been completed, and it puts Andrew Wilkie 1091 votes (1.0 per cent) clear of Labor. That’s a big hurdle for Labor to clear on absents and postals, but there are too many imponderables to say it can’t happen.

Monday, August 23

11pm. “Only one seat now in doubt as Wilkie loses bid for Denison”, reports the Sydney Morning Herald, and it’s probably not alone. This misapprehension is based on the ABC computer’s projection of the Labor-versus-Wilkie indicative preference count, which assumes the 20 booths that haven’t been counted will follow the preference pattern of the 26 that have. There is a three-sided problem here: Labor’s share of the preferences is not as high in areas where they are weak generally; the booths are being counted in alphabetical order; and the strongest Liberal booths begin with an S. Antony Green’s modelling to account for this turns the projected 0.6 per cent Labor lead into a 1.1 per cent deficit (subject to a margin of error), a view shared by PB commenters who know their way around a linear regression. However, Labor is likely to at least close that a little on postal votes.

6pm. Labor might appear to have the advantage superficially at present, but Sykesie in comments has produced a model accounting for the association between Labor’s primary vote in booths that have reported and the share of their preferences in them. The upshot is that as counts are added for booths less preferable to Labor are added, their share of the preferences will come down, Sykesie projects them to finish on 48.4 per cent with an error margin of only 1.3 per cent. However, that doesn’t factor in the likelihood that Labor’s position will improve as postal votes come in. That still makes it too close to call, but Wilkie would probably be favoured.

2.30pm. The Electoral Commission is conducting a thrilling indicative preference count between Wilkie and Labor to ascertain what will happen if they are indeed the final candidates. Wilkie currently looks to be just slightly under the share of preferences he needs, but it’s been back and forth as booths have been progressively added in alphabetical order.

Sunday, August 22

Accomplished Tasmanian psephologist Kevin Bonham, who closely observed the behaviour of Greens preferences in relation to Wilkie when the latter ran at the March state elections, disagrees with Possum’s assessment that Greens preferences will not necessarily put Wilkie ahead of the Liberals, and thinks a greater threat to Wilkie would be that he might be overtaken by the Greens, who will have run a better resourced postal vote campaign. If he’s right, the surprises in Denison might not be over. It is mostly being taken for granted that Liberal preferences will allow Wilkie to ride home over Labor if he finishes ahead of them, but a WA Labor source advises caution on this count based on the precedent of Kwinana at the September 2008 state election. It was widely thought after election night the seat had been won by independent Carol Adams, but victory slipped away from her due to the surprisingly high number of Liberal voters who had Labor second.

Saturday, August 21

This post will be used to follow the late count in Denison, where independent Andrew Wilkie superficially looks well placed to win a Labor seat vacated by Duncan Kerr and contested for them by Jonathan Jackson. At issue is the distribution of preferences from the fourth-placed Greens, who polled 19.01 per cent. Wilkie needs them to close what at present is only a 0.1 per cent deficit over the Liberals, but Possum at least believes the fact preferences are splitting three ways between Labor as well as Liberal and Wilkie will land him short, especially after factoring in a likely weakening of his position as postal votes come in. However, the ABC reports “Labor scrutineers are predicting a desperately close result as preferences are distributed”.

Photo finishes: Lindsay

Friday, August 27

3pm. 968 postals favour Labor 518-450, lead out to 1508.

Thursday, August 26

5pm. 1373 postals have also favoured Labor 774-599: Labor now looking home and hosed with a lead of 1442.

3pm. 955 postal votes added, and they’ve surprised slightly by favouring Labor 486-469. Labor now leads by 1267, and my projection says much the same.

Wednesday, August 25

11pm. 1887 postal votes have broken 1005-882 to Labor, increasing the margin to 1250.

1pm. They seem to be taking their time with this one. No non-ordinary votes added yet, but rechecking of ordinary votes has added 131 to the Labor lead, which is now at 1136.

Saturday, August 21

This post will be progressively updated to follow the progress of late counting in Lindsay, where Labor member David Bradbury finished the night with a 1012 vote (0.69 per cent) lead over Liberal challenger Fiona Scott. This lead narrows to 0.2 per cent on Antony Green’s projection.

Photo finishes: Corangamite

Saturday, August 28

Corangamite is back on the AEC undecided seats list (a list of one), but given not much has actually changed, this suggests the current margin is right on their dividing line. 1877 abent pre-polls have gone 996-881 the way of the Liberals, but 1299 absents have favoured Labor 697-602: a net gain for the Liberals of only 20 votes. Labor leads by 884, while my projection has it at 644.

Thursday, August 26

6pm. Absent votes continue to flood Labor’s way, a batch of 924 going 557-367 and blowing the lead out to 895.

12noon. The first 938 absent votes have been added and they’re a good batch for Labor, breaking 535-403 in their favour and widening the lead from 573 to 705. However, I’m told Labor are worried about what the remainder of the absent count might have in store – much of it might come from the snowfields and Queensland, producing a different absent vote trend from what might be expected at a summer election. The expectation was that Labor would still slip over the line, but only just. I was told this before this present batch of absents was added – possibly it has soothed nerves a little.

Wednesday, August 25

2pm. Another 1466 postal votes have been added, and while they have split 766-700 in favour of Liberal and narrowed the Labor lead to 571, they have reduced the overall Liberal share of the postal votes counted from 57.67 per cent to 56.19 per cent. This means my projection of the final Labor margin has increased from 240 to 315.

Tuesday, August 24

11pm. I had not previously been considering absent pre-polls, which unlike ones cast within the division remain to be counted. The AEC says there are 3508 of these, and if they behave as pre-polls overall did in 2007 my projection of Labor’s winning margin narrows to 240.

7pm. Another 1960 postal votes haven’t favoured the Liberals to quite the same extent, although they have narrowed the margin in absolute terms by 280 to 637. The Liberal share of the postal vote has dropped from 58.2 per cent to 57.7 per cent.

4pm. A surprise from the first batch of postals, with 1955 votes favouring the Liberals even more heavily than in 2007: 1134-821, closing the gap from 1237 to 917. If the trend continues, the gap could narrow by maybe another 700 or 800. However, absent votes are likely to favour Labor, and none have been added yet.

Sunday, August 22

Special hospital team votes have cut 99 votes from Cheeseman’s lead, now at 1189 (0.76 per cent).

Saturday, August 21

This post will be progressively updated to follow late counting in Corangamite, where Labor member Darren Cheeseman finishes the night 1288 votes (0.83 per cent) ahead of Liberal challenger Sarah Henderson. Owing to the extraordinarily strong performance by defeated Liberal member Stewart McArthur on postal votes in 2007, which is used as the benchmark for the projection, the ABC computer has it at 0.0 per cent.

Photo finishes: Hasluck

Friday, August 26

477 pre-polls go 252-225 to Liberal, lead out to 868.

Thursday, August 26

5pm. Another 806 postals go 447-359 to Liberal; the lead is now 820.

3pm. 1865 absent votes have favoured Liberal 946-919, increasing the lead to 721.

Wednesday, August 25

5pm. The Liberals have enjoyed a good batch of postals breaking 801-616 their way, increasing their postal vote share from 52.7 per cent to 54.1 per cent. They have also gained five votes from 667 pre-polls and 38 from further rechecking of ordinary votes. That increases the margin to 770 votes, though my projection is that outstanding absent votes will rein that in to 369.

2pm. Rechecking of ordinary votes have boosted Labor by 99 votes, while a further 955 absent pre-polls have followed the earlier trend in favouring the Liberals 505-450. Extrapolating current swings on pre-polls, postals and absents across uncounted votes, I project a Liberal margin of 269 before taking into account the small but unpredictable share of provisional votes, on which Labor could maybe hope to recover as much as 100 votes if they’re lucky.

Tuesday, August 24

7pm. The Liberals have gained another 131 votes through rechecking of ordinary votes, which have produced a net gain of 80, and a further 705 pre-polls, which have added 51. The margin is now 586.

4pm. 2546 postal votes have been added out of what will be a total of about 3500 and they haven’t delivered the miracle Labor had hoped for, favouring the Liberals 1775-1417 and pushing the lead out to 455. This is still quite a lot better for Labor than in 2007, when the Liberals’ postal two-party vote share was 7.1 per cent higher than for ordinary votes rather than 2.5 per cent. Funnily enough, the current Liberal lead is exactly what Labor were able to scrounge back on absent votes in 2007. However, that becomes 330 when you factor in the 1 per cent swing against Labor, and the remaining postals look likely to widen the gap by maybe another 100 votes. But could provisional votes hold an ace for Labor? Last time there were only 130 of them compared with 638 in 2004, because self-serving Howard government law changes made votes cast by people who had moved within the electorate without updating their address inadmissible. This time provisional votes will include the 600 or so late-enrolling voters who were enfranchised by the High Court ruling in favour of GetUp!, nearly every one of whom has been personally contacted not only by the AEC, but also the ALP who know such votes would heavily favour them. (UPDATE: Propmanoz in comments explains I was wrong about this: supplementary lists of affected voters were used to admit the voters on polling day. The only provisional vote dividend for Labor would be an increase in the number of voters who make the effort to follow up on providing the required identification if they had none on the day, which would be expected given the importance of the contest).

3pm. 704 absent votes have broken 376-328 Labor’s way – possible the Gosnells area I affect I alluded to two days ago – which has narrowed the Liberal lead from 363 to 317. That makes the current batch 3.7 per cent better for Labor than the ordinary votes compared with 2.2 per cent in 2010, so there’s a suspicion the rest of the absent count will come from different areas and thus not favour Labor to the same degree. There should be at least 2000 absent votes to go: if the trend from the first batch continued, Labor would claw back an extra 136 votes.

Monday, August 23

921 absent pre-polls, cast at centres outside the electorate an thus not counted on election night, have gone 477-444 in favour of the Liberals and widening the lead to 382. Contrary to what I foreshadowed yesterday, no polling day absent votes have been added.

Sunday, August 22

Tomorrow there will be about 1000 absent votes added, which Labor expects to be dominated by Gosnells area residents voting across the boundary at the Westfield Carousel megaplex – a Labor crowd. Absent votes from farther-flung locations will follow over coming days. A bagful are expected from Broome where, I am told, 50,000 people are attending a festival (I think it must be this one, notwithstanding that it does not actually start for a few days). I would presume its audience to be “blue-green”: resident of the prosperous hills suburbs around Kalamunda, but not entirely representative of them. Given there were 6500 absent votes in 2007, that still leaves a lot left over from various other locations.

Throughout and beyond, postal votes – perhaps as many as 4000 – will be added to the count, in addition to a smaller number of international votes as they arrive. This is where Labor may have an ace up its sleeve: it had 2420 applications processed compared with just 681 for the Liberals, dramatically reversing a 911-549 deficit in 2007. This gives them the advantage that they and not their opponents can target their residences with timely campaign and how-to-vote material (an area self-evidently crying out for reform).

Saturday, August 21

This post will be progressively updated to follow late counting in Hasluck, where Liberal candidate Ken Wyatt finishes the night with a lead of 369 votes (0.28 per cent) over Labor incumbent Sharryn Jackson. This turns into a 0.6 per cent lead on Antony Green’s projection.

Photo finishes: Brisbane

Friday, August 27

Rechecking of ordinary votes has cost Labor 87, and for the first time a batch of absent votes has favoured the LNP, albeit slightly – 622-613. LNP lead now 839, which should quite comfortably be enough.

Thursday, August 26

7pm. Another 1145 absent votes have gone 599-546 to Labor: very slightly good for them in absolute terms, reducing the margin to 743, but Labor’s 52.3 per cent share compares with 56.5 per cent in the first batch. A better portent for Labor with Antony Green apparent confirmation that today’s LNP-friendly batch of postal votes came from the Enoggera army barracks.

4.30pm. The LNP have moved another 112 ahead with the first 596 pre-polls breaking 354-242 their way: margin now out to 796.

3pm. Lucky I advised that caution yesterday, because a second batch of postals has been wildly unlike the first, favouring the LNP 1686-1384. This has widened the lead from 382 to 684 and slashed the overall ALP share of postals from 56.4 per cent to 47.8 per cent. My projection now has the LNP prevailing by 403, and I suspect that might flatter Labor with respect to the absent votes which are probably unlikely to continue splitting 56.3-43.5 to Labor. Pre-polls remain a wild card, but the LNP are back in the box seat.

Wednesday, August 25

11pm. Hmm. The first batch of 961 postal votes has heavily favoured Labor by 542 to 419, cutting the margin to 382. This is a swing in Labor’s favour of 0.8 per cent against the postal vote total in 2007, compared with a 4.9 per cent swing the other way on ordinary votes. Two questions arise: will this prove typical, and how many postal votes will there be? The scrutiny progress sheet says 602, which is obviously wrong because 961 have just been added to the count and a more typical amount is 5000 (UPDATE: And I now note that informed sources in comments are saying to expect exactly that). If you project Labor’s share of the first batch over the latter figure – which still seems intuitively optimistic from Labor’s point of view – they end up in front. If the 5000 outstanding absent votes also follow the trend of those already counted – ditto – Labor ends up winning quite handily, by about 500 votes. Caution advised at this stage.

5pm. Absent votes continue to flow heavily to Labor, raising their hopes of a late boilover. 1182 were added today, breaking 671-511 in favour of Arch Bevis and reducing his deficit from 657 to 490. My projection is for an LNP win by just 195 votes, but this is based on the AEC’s figure of 602 postal vote envelopes issued which is surely much too low. If you assume about the same many postal votes as last time (about 5500), the LNP winning margin blows out to about 400.

Tuesday, August 24

The first 978 absent votes have split 549-429 in favour of Labor. Comparing relativities of ordinary and absent votes from 2007 and 2010, this is a reversal of nearly 10 per cent. What this probably tells us is that these absent votes are from booths in Labor-leaning areas slightly outside the electorate’s boundaries, and the remaining 5000 or so absent voters are unlikely to be as favourable for them.

Saturday, August 21

This post will be progressively updated to follow late counting in Brisbane, where Liberal candidate Teresa Gambaro finishes the night with a lead of 858 votes (0.68 per cent). This narrows to 0.2 per cent on Antony Green’s projection.

Photo finishes: the Senate

Saturday, September 11

As you may have noticed, I have kind of dropped the ball on following the late Senate count – normally a richly absorbing pastime in the post-election dead zone, but in the past three weeks there have been bigger fish to fry than the precise make-up of a Senate that will clearly have the Greens holding the balance of power. The AEC has pushed the button on the counts in Queensland (three LNP, two Labor, one Greens) and the ACT (one Labor, one Liberal), with the others presumably to follow shortly. There has never been any doubt about New South Wales (two Liberal and one National, two Labor, one Greens), Western Australia (three Liberal, two Labor, one Greens), Tasmania (three Labor, two Liberal, one Greens) and the Northern Territory (one Labor, one Country Liberal Party). Doubt has also faded about the remainder:

Victoria. What at first appeared the quirky prospect of win for DLP candidate John Madigan has firmed. I personally anticipated the resources of incumbency might mean Family First would perform strongly on declaration votes, but they have actually gone backwards: according to the ABC projection, a 0.12 per cent deficit against the DLP at the revelant point in the count on election night has actually widened to 0.22 per cent. The Coalition have also failed to make up the ground needed to overtake the DLP at the penultimate count, with the gap now at an unassailable 0.72 per cent. That leaves us looking at a result of two Labor, one Liberal and one Nationals, one Greens and one DLP.

South Australia. There was briefly the prospect a week ago of an upset win by Bob Day of Family First at the expense of third Liberal candidate David Fawcett, but we now appear to be looking at a vanilla-flavoured three Liberal, two Labor and one Greens result. At the second last count, the ABC computer projects Fawcett to be on 8.89 per cent against 8.16 per cent for Day. With Day excluded, Family First and other right-wing preferences put Fawcett well ahead of the third Labor candidate.

There is thus little doubt that the newly elected Senators will include 18 from the Coalition, (12 Liberal, two Nationals, three Liberal National Party and one Country Liberal Party), 15 from Labor, six from the Greens and one DLP. These will join the state Senators whose terms began following the 2007 election – 16 Labor, 16 Coalition (14 Liberals and two Nationals), three Greens and Nick Xenophon – for a total result of 34 Coalition (26 Liberal, four Nationals, another three from the Liberal National Party including two who will sit with the Liberals and one who will sit with the Nationals, and one Country Liberal Party who sits with the Nationals), 31 Labor, nine Greens, one DLP and Nick Xenophon.

Monday, August 23

Kevin Bonham and GhostWhoVotes in comments note my assessment of Tasmania was mistaken, as it wrongly allocated Liberal preferences to the Greens over Labor like they normally would. There is in fact little doubt the final result will be three Labor, two Liberal and one Greens, meaning the defeat of Liberal incumbent Guy Barnett by Labor’s Lisa Singh, a former state government minister who lost her seat in Denison at the March state election.

Saturday, August 21

A brief and bleary summary of the Senate situation, based on a quick and dirty review of Antony Green’s projections. This will be progressively updated as further results come to hand. The Greens look good for a Senate seat from each state and will hold an unassailable balance-of-power position in the Senate. The best shot for a quirky result is Victoria where the Democratic Labor Party are currently in the hunt – as is incumbent Steve Fielding, despite reports to the contrary.

New South Wales. This looks like a reasonably straightforward result of three Coalition and two Labor, with Lee Rhiannon of the Greens on a 2.5 per cent lead over the third Labor candidate for the final place.

Victoria. It is widely being reported that Steve Fielding has lost his seat, based on assessments of Antony Green’s projection that went no deeper than the predicted final result. The remarkable fact of said projection is the win for the Democratic Labor Party, who would be advised not to count their chickens. Four counts earlier, the DLP emerge ahead of Steve Fielding by the narrowest of margins, resulting in the former receiving the latter’s preferences and vice-versa. The DLP then emerges ahead of the third Coalition candidate and wins the seat on their third preferences, but it could just as easily be fielding who does this. Alternatively, neither could win – Fielding or the DLP could fail to get ahead of the third Coalition candidate, who might end up taking the seat instead. Or they could get ahead, but then fall short of overtaking Labor in the final count, so that Labor wins the seat.

Queensland. Three Liberal National, two Labor, one Greens.

Western Australia. A delightfully straightforward result, with the Liberals just over three quotas, Labor just over two (a remarkably low 29.8 per cent) and the Greens almost bang on one.

South Australia. Bob Day of Family First looks like he’s come close to overtaking the third Liberal candidate, but is currently 0.4 per cent behind and likely to lose ground in late counting. That being so, the final seat looks set go to the third Liberal, who looks about 3 per cent ahead of third Labor. Result: three Liberal, two Labor, one Greens.

Tasmania. Not only has Christine Milne been easily re-elected for the Greens, the current ABC projection has their second candidate just 1 per cent short of overtaking the Liberals at the second last count, and then winning the seat at the expense of a third Labor candidate. The exclusion of the second Green instead delivers the latter a narrow win over the third Liberal candidate. However, the unusually high number of below-the-line votes in Tasmania might makes things unpredictable. Realistically, the contest is between Labor and the Liberals to take a third seat, with the former slightly ahead.

The territories. Liberal Gary Humphries has only just cleared a quota (one third) in the ACT, but will be made comfortable by Democrats preference and a high rate of leakage. Equally, Labor wasn’t too far over a quota in the Northern Territory.