Sunday, November 1, 2020

Queensland 2020 Postcount

Labor re-elected with increased majority

Seat total after recounts ALP 52 LNP 34 KAP 3 GRN 2 PHON 1 IND 1

Labor won Bundaberg and Nicklin after recounts.

LNP has won Currumbin

----------------------------------------------------------------------------------------------------------

Donations welcome!

If you find my coverage useful please consider donating to support the large amount of time I spend working on this site.  Donations can be made by the Paypal button in the sidebar or email me via the address in my profile for my account details.  Especially in these uncertain times, please only donate if you are sure you can afford to do so.

---------------------------------------------------------------------------------------------------------


Intro (Sunday)

Welcome to a shorter than expected Queensland postcount thread.  Labor has been returned with what looks like a 2PP swing of around 2% (maybe more) in an election where very few seats seem to be changing hands.  At the moment Labor has lost South Brisbane to the Greens but is not close to losing to the LNP anywhere.  The LNP has lost Pumicestone and Caloundra and appears to have lost Hervey Bay on a monster swing. All these were vacant seats.  A few of its incumbents are at some risk.  

I will be unrolling seats here progressively through the day.  There are a number of classic-2PP seats where the LNP holds leads of 51-49 or over and where the ABC was not projecting this lead to close to below 51-49; these are Burleigh, Chatsworth, Clayfield, Coomera and Glass House.  I am assuming that the LNP will win all these but I will cover them in detail should figures change in a way that places them in significant doubt.  I am also for the time being not covering Cooper in any detail.  Cooper has seen a large swing to the Greens but at the moment they are 2.1% behind the LNP for second place on primary votes, and the postal vote count is not very advanced in Cooper, so I cannot see the Greens bridging that gap (which I expect to widen).  

Regarding Cook, the 2CP between Labor and KAP is very bouncy because of the huge variation between booths in this seat, with Labor dropping back to 51.8% today.  It is currently very doubtful that KAP even get into second, but if the count shifts in KAP's favour I will cover that seat in detail.  

A note about South Brisbane: Jackie Trad's share of all preferences vs the Greens (currently running at 35%) is very weak.  This suggests to me that she probably would have lost even had the LNP preferenced her, although it is plausible that she would have just won.  There are two possible explanations here; one is that the LNP how-to-vote card was more strongly followed than expected (usually Labor gets about 40% of preferences in this situation) and the other is that Trad's preference share was dragged down by the LNP's attacks on Trad, and this would also have affected her had she been preferenced.  So I am somewhat sceptical that the LNP's decision determined the outcome, though it certainly destroyed any chance Trad might have had.  

Each seat below has its own section where updates will be posted until the seat is no longer of interest to me.

Notes About The Counts

Some general postcount themes: in this election a lot of the remaining action will be out of electorate prepolls.  Out of electorate votes on the day (absents), which tend to skew to the left, will be a smaller proportion than normal because voting on the day is down.  The extent of uncounted postal voting varies by seat and some seats have large prepoll centres still to report.  The counting so far is very rough and there may be significant count errors that will place unexpected seats into play.  

The ECQ's "Unofficial Indicative Count"s were initially not being updated beyond the first night, but now the 2CP counts are being updated as new votes are added.   

Bundaberg (LNP, 4.2%)

Bundaberg was among Labor's more preposterous gains in the 2015 dumping of the one-term Newman government, the party just winning it back with a 19.8% swing.  It was considered pretty much undefendable in 2017 and was the only Labor-occupied seat lost to the LNP.   It wasn't really on the radar in 2020 either until just in the last few days the LNP's south-east coastal malaise was seen to be spreading northwards and a view emerged that it might fall.  Early in the night Bundaberg was projected by the ABC as a comfortable LNP hold, but late in the night the projection changed as prepolls favoured Labor.  (Contrary to some of my late-night comments, the 5.5% vote for Legalise Cannabis probably doesn't have a huge amount to do with it, as the LNP is still getting 56% of all preferences.)

Labor's Tom Smith leads LNP incumbent David Batt by 277 votes (50.60%).  The ABC projected this to widen to 51.1%.  I project about half the lead to disappear, perhaps slightly more, on 2833 postals not yet counted to 2PP.  The remaining count might be around 4800 votes including around 2800 postals, the rest being absents, out of electorate prepolls and declaration votes.  I don't think this will blow out in the way the ABC was projecting it to but at the moment Labor seems reasonably placed.  

Monday morning: The ABC currently projects Smith 181 votes ahead (50.35)

Monday night: To the wire for now with just 24 votes in it.  In Smith's favour, absents have yet to be counted, and out-of-electorate prepolls are not included yet either.  

Tuesday 4:30: Both the ABC and the ECQ now have Batt in front, by 4 and 3 votes respectively.  A very high 86.5% of postals issued here have been returned (I have generally been assuming a 75% return rate) and this has probably interfered with the ABC's earlier projection.  However (i) this probably means very few postals to come (ii) there are still no out-of-electorate prepolls counted, so there is quite a lot of scope for Labor to come back in the rest of the count. (However absents have been included contrary to earlier comments, with Batt winning today's by 28).  

Wednesday 4:00: Labor now 12 votes ahead (ABC) or 9 votes ahead (ECQ).  

Wednesday 7:10: The ABC projects Batt 48 votes ahead while the 2PP count appears to have disappeared from the ECQ site.

Thursday 4:00: The ABC projects Batt 86 votes in front.  

Monday: The ABC projects Batt 112 in front.  Not much to do here except wait for the preference throw.

Wednesday: The ABC projects Batt 106 in front, so I now expect Batt to win.  Note that InDaily published estimates that were much larger than the ABC's for all seats, but I am unaware of the basis for those.  

Wednesday night: Labor reports Tom Smith has won Bundaberg! A report that the margin is 11 votes so I assume this is pending a recount.  

Thursday: Bundaberg will be recounted.  11 votes is close enough for a realistic chance of a different winner.  

Friday: Labor wins by nine votes!  This may go to court, but not necessarily (Fisher by-election SA 2014 had the same margin and was not taken to court.)  Anyway Smith will be seated in the meantime.

Currumbin (LNP, 3.3%) - LNP won

Currumbin was the scene of a better-than-average performance by Labor in a by-election earlier this year after long-term incumbent Jann Stuckey quit in acrimonious circumstances.  The LNP retained the seat with a swing against it, leaving new MP Laura Gerber with only several months to establish herself before the by-election.  As such, the seat was at heightened risk.  

Gerber leads Labor's Kaylee Campradt (her opponent in the by-election too) by 268 votes (50.54%).  It's rare for conservatives who lead by this much at the end of the first night of a count to lose, and at the moment I think Gerber will retain.  However, the postal count is quite advanced in this seat with 3707 counted; I estimate 2600 to go, at least some of which will be later postals (which tend to be weaker for the LNP).  The remainder, which I estimate at 4700, will mostly be out of electorate prepolls and absents.  Maybe the postals will put it beyond doubt or the out of electorate prepolls will do nothing or favour the LNP, but I think this is close enough on the existing count to regard as in more doubt for now than the others where the LNP is leading.  

Monday 1:20  The ABC now shows the gap as down to 141 votes - I believe on rechecking as this is without significant changes to the overall total.

Monday 2:10 The ABC has now reverted to its previous numbers.  My estimate from patching together the two counts is Gerber is 145 votes ahead.  

Monday night: Gerber 265 ahead.  Over 500 absents have been added, which made next to no difference, but I don't know where they were from.

Tuesday 6:30: Gerber 302 ahead, but I project her to gain another 50 from out-of-electorate prepolls that have not yet been thrown to 2PP.  Extremely hard for Labor to win from here and I am treating this seat as an expected LNP retain (but keeping an eye on it).

Wednesday 4:50: Campradt has done better on the out-of-electorate prepolls than I expected (possibly because more were added; haven't checked) but is still 265 behind.

Wednesday 7:15: Well, I have to put this one back into the seriously doubtful category (that's why I usually just say I expect someone to win them rather than "called"), as the gap is down to 43 votes according to the ABC. The cause of this change is unclear to me, as the earlier figures with Labor 265 behind were off slightly more (!) primary votes.  Perhaps a correction?  Or perhaps the ABC projection is incorrect? Investigating ...

Thursday 11:00: I have had a look at what is going on in this count following suggestions on Twitter that the ABC figures, which narrowed yesterday, are wrong.  Antony Green has tweeted "The narrowing was caused by the ECQ stopping the feed of preferences which caused estimates to kick in." I have used the Poll Bludger count, which presently shows the LNP ahead by 227 votes, as a base, and compared it to the official live count.  All candidates have lower primary vote totals in the official live count - particularly Gerber drops 52 votes on Absent Early Voting -  and when I apply a record of the preferences that I kept earlier in the count this drops the LNP's lead to an estimated 172 votes.  However if I apply the preferences I previously recorded to the whole of the current figures, the LNP's projected lead drops to 50 votes.  

Thursday 4:00: The ABC projects Gerber's lead stretching back to 109 votes. 

Friday 4:45: The ABC projects Gerber's lead now at 142 votes.  

Wednesday: The ABC projects Gerber 145 votes ahead, so I now expect Gerber to win.

Wednesday night: Gerber has won by 310 votes according to Campradt - there will not be a recount. 

Late Wednesday night: That one's over, Currumbin declared.

Nicklin (LNP, 5.3%)

Another seat that was not really on the radar because it lay on a higher margin than a bunch of Gold and Sunshine Coast seats that the LNP has defended - and also because Labor has never won it.  It was, however, held for a long time by independent Peter Wellington.  Currently Labor's Robert Skelton leads one-term LNP incumbent Marty Hunt by a precarious 82 votes and this looks like something that could easily fall over.  The count in Nicklin is very incomplete at 61.6%, and I think out of electorate prepolls (likely to weakly favour Labor) will be more of the rest than postals (of which I estimate 3200 to come).  The ABC was projecting Labor to hang on; we'll have to wait and see how plausible that is. 

Monday 1:20: ABC projects gap down to 66 votes, presumably on rechecking.

Tuesday 7:00: Labor won absents today while the LNP has so far narrowly won out-of-division prepolls.  The current gap is 67 with 76.2% counted.  I believe the vast majority of the rest here will be postals and that Labor would be very nervous about those.  

Wednesday 1:00: The gap has closed to 30 votes with 82.4% counted.  I project it to close to about 15 votes when votes not yet available to 2PP are thrown.  5902 postals are gone; I expect there are still at least several hundred to come; the theoretical maximum is 2778.

Wednesday 7:10: Skelton now shown as a handy 112 ahead by the ABC and 109 ahead by Poll Bludger.  Still only 81.9% counted but that gain, if correct, has improved Labor's chances, as postals probably won't now knock it over by themselves.

Thursday 4:00: The ABC projection of Skelton's lead is out to a whopping 206 votes!  If that's accurate it will be very difficult for the LNP to recover.  

Friday 6:20: With the ABC's estimate of Skelton's lead out to 259 and probably not more than hundreds of postals left to add, I am now upgrading Nicklin to expected win status.

Wednesday: The ABC projects Skelton 150 votes ahead.  

Wednesday evening: Labor say they have won Nicklin.  The margin is 79 votes which may be recounted, but is very unlikely to be overturned.  I cannot find clear information on whether or not the ECQ has an automatic recount threshhold, but Bulimba 2012 was recounted on a similar margin (the margin reportedly changing from 85 to 75).  

Thursday: Nicklin will be recounted.

Friday: The ABC is showing the gap in Nicklin at 39 votes despite using the same primary vote figures as the ECQ.  The reason for this is unclear to me at present.

Friday evening: Labor wins Nicklin by 85 votes after recount.

McConnel (ALP vs Green, 2.9% 3CP) - ALP won

I am very confident Grace Grace has retained McConnel but it is an unusual case that is worth keeping an eye on.  In 2017 the leading primaries in McConnel were LNP 36.5 Labor 33.7 Greens 27.1.  The current primaries are Labor 35.7 LNP 30.9 Greens 28.2.   The Greens have failed to outpoll Labor, but if they could get in front of the LNP they would win on LNP preferences.  The postal count is fairly well advanced in McConnel but there may still be around 4000 formal postals to come.  If they are the same as the formal postals so far, they would more than double the LNP's lead, but I suspect the damage will be less.  However, the Greens are not even gaining on the LNP based on out-of-division Brisbane CBD prepolls so it is not easy to see where they would make large gains to counteract the existing lead plus postals.  In 2017 there were a lot of absents in McConnel (over 2400) but those only favoured the Greens relative to the LNP by 3.1 points overall, which would do nothing here, and there would be fewer of them anyway.  The Greens may well gain 1-1.5% or so on the LNP on preferences from Legalise Cannabis and an independent but I suspect they will be too far behind for that to help. 

Monday night: Gap stretches to 3.39%.

Thursday night: Gap down to 2.64%, but it might not change much from here on.  The Greens have done pretty well to not be further behind at this stage.

Wednesday: And now we know the Greens did not get into second as ... the seat has been declared! (Primary vote gap was 2.82%)

12 comments:

  1. A factor in Bundaberg and Hervey Bay, the unfair compulsory cashless debit card rolled out by the Federal LNP

    ReplyDelete
  2. According to the ECQ site, the legalise cannabis preferences slightly favoured the LNP in Bundaberg. A lot of this would have to do with ballot paper order. Not necessarily donkey votes as such, they may well consciously opted for that party, but then just gone straight down the ballot after that. Since the LNP guy was higher, he almost certainly has that to thank if he survives -just as the ALP woman did in Eden-Monaro

    ReplyDelete
  3. Is Andrew home in Mirani? I thought postcounts were generally bad for non Green minor parties. How far behind are LNP at exclusion?

    ReplyDelete
    Replies
    1. Currently LNP are 5.5% behind on primaries but we don't know how far behind they are after preferences as there has been no 3CP throw. So far they've been catching Andrew on postals and out-of-electorate prepolls at a rate approaching 8% of votes counted, but (i) the prepolls might be from somewhere unrepresentative (ii) late postals are weaker for LNP than early postals - last time the LNP ended up only beating him by about 2% on all postals (iii) last time he did well on absents (iv) even if that rate projects across the whole postcount he should still be 1% ahead on primaries and I doubt the LNP would catch that. I'd say in all probability he's fine but the rate of gain from the LNP so far is worth keeping an eye on.

      Delete
    2. ABC now showing Mirani count at 74.4% complete with LNP still 5.3% behind. This seems to be on account of a massive addition of Absent Early Voting of which over 7000 are now showing counted by ECQ in the unofficial count.

      Delete
  4. Re:Bundaberg. It's election day absents that have put Batt into the lead. He won today's batch by 28.

    ReplyDelete
  5. I heard that the 3PP in McConnel was
    GRN 9,939 (30.2%)
    LNP 10,855 (33.1%)
    ALP 12,044 (36.7%)

    I guess the One Nation preferences were as bad as the Cannabis were good for the Greens vs LNP.

    ReplyDelete