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”.
haha the AEC web site says 99.4% counted! surely an error
17:41 99.4 Denison ALP 15.3 52.9 12.4% from ALP ALP RETAIN
Nope just stopped for dinner.
Hobart South prefs 70% to Wilkie
Hobart West 67%
– so neither side is pulling away
If Labor is polling all 13000 pre-polls, Denison must have turned into a dictatorship!!!
And when the absentee and postals report, it could have more than a 100% turnout!!!
Fantastic thread guys.
Assuming that Labor ends up a little under 49% on the “normal” TPP, and Kevin’s conditions (1) and (2) are also satisfied, does anyone have any idea what sort of postals campaign Wilkie ran?
[According to the site, theres been 13,136 early votes cast in Denison, which is absolutley plausible.]
I don’t think the true prepoll number is anything like that. There were 4646 prepolls in Denison last time. Looks like it’s just some glitch causing that number to appear in the wrong place. Sheesh, as if there isn’t enough to confuse the MSM in this count!
[Assuming that Labor ends up a little under 49% on the “normal” TPP, and Kevin’s conditions (1) and (2) are also satisfied, does anyone have any idea what sort of postals campaign Wilkie ran?]
None whatsoever, which, apparently, is not much less than Labor. See Laocoon’s excellent data in #4.
I believe he also ran none in the state campaign where his primary on postals was only 0.8 points below his primary on other votes.
Thanks Kevin. Very interesting.
After 24 booths the model is as follows:
(useful to compare to my earlier model at post #65 which was after 15 booths):
ALP 2PP% = 1.143*(ALP primary) + 0.390*(Green primary)
error in coefficients: (1.143+/- 0.010); (0.390+/- 0.017)
standard error of prediction: 1.40%
Prediction: Labor 2PP of 48.5%.
The equation and coefficients havent changed alot from post #65, although the error
has increased slightly from 1.18% to 1.40%, which in a close race might be quite important.
The equation depends heavily on the labor primary vote, so labors 2PP tanks quite quickly as the labor primary vote drops.
There must be some pretty bad booths to come for labor.
For what its worth – I hope this model fails and labor win Denison!
Going out for a few hours now.
Thanks Kevin – I’ve enjoyed reading your contributions. I’m off to cook pasta.
[There must be some pretty bad booths to come for labor.]
Ooooh yeah. The blue-blood booths of toxic doom, the last enclaves of Liberal supremacy in Denison, await JJ and it’s not going to be pretty. 😉
Oh no, Kevin. You can’t go out!
Can’t you have some food sent in?
I now appreciate my ignorance.
4 crappy booths added which significantly shrink the labor lead.
Ironically, this has improved the model slightly for labor, now predicting a final 2PP of 48.6% on a 1.46% error margin.
sykesie, top work.
how much do the relative populations of each booth distort your model? I notice that while 28/56 booths are now counted, significantly more than 50% of the votes have – disregarding pre-postals etc.
An excellent question Traveller.
I actually gave this some thought an hour or so ago and probably need to think about scaling the results in some way. However, with 28 data points now in the model, I dont believe scaling for number of votes per data point should affect the model too much … I think!
Antony Green on Twitter is just now “projecting” Wilkie on 51.1%
Well, it must offer some semblance of reassurance that we’ve knocked over half the booths, but 57% of the votes. Yeah?
Thats probably true – but you’d need to carefully check and hope there are some big pro-labor booths that might skew the outcome more towards labor. The next 4 booths are better for labor so expect to see the lead grow again.
There are some very good booths for the ALP in the last 24 physical booths (about 8). A few with over 50% Primaries.
However, there are also plenty of shockers!
The Sandy Bay ones are particularly nasty.
I am sure there is a statistical reason why this is flawed BUT – here is a possible method of scaling.
The total votes so far distributed = 31,878
Total ALP primaries in the distribution = 11,934 or 37.43%
Total ALP primaries = 36%
Therefore this sample deviates from a true sample by up to 1.43/36 = 4% or 1,275 votes. I think Labor will be within the suggested 49% before postals ( but personally I think the postal advantage will be greater)
actually redoing that it ends up with labor at 48.6 – same as the model
That’s spooky … it must be time for bed 🙂
[Oh no, Kevin. You can’t go out!
Can’t you have some food sent in?]
I think eating at the chessboard during my tournament game would have been discouraged.
I see Jackson is still notionally ahead on 2PP but there are 12 Hobart booths and only 7 Glenorchy booths to go and some of those Hobart booths are like LSB, ie particularly evil for anyone who has ever had anything to do with the Labor Party so Wilkie should be at 51-point-something. What the something is and whether Jackson can catch it on postals will be interesting to see. Note that it’s not just Jackson’s performance on postals that matters, but also, for instance, Couser’s. If Couser goes up at the expense of Simpkins then that’s a weaker preference flow on those votes.
The remaining booths are not only relatively poor for Jackson but they are also relatively good for Simpkins which in turn is good for Wilkie.
I assume they stopped for the night long ago.
[there are 12 Hobart booths and only 7 Glenorchy booths to go]
Correction, it’s 11-8, the vote counts being 11210-7555.
I know others have said it but this post really shows just why the main-stream-media is failing. Kevin’s local knowledge (the flannelet curtain!) enables us to get a far better picture of what’s really happening than we will ever get from an hysterical Shanahan, a dreary Grattan or a flatulent Kelly.
#129
It’s disappointing, but to be fair you can’t expect them all to be across the nuances and quirks of the seat of Denison, and the odd fact that all the better Labor booths seem to come first alphabetically.
I note the AEC’s Virtual Tallyroom site now has a big notice to inform people that the numbers are not final yet and close seats could still change hands.
I see the ALPs progressive count lead has widened a tiouch this morning – still pre-Sandy bay though…
Kevin or anyone – would I be right that Tolosa looks good for ALP if they’re still in it after sandy bay count?
@132
At 50.5% of primary vote I think we can safely say “yes”.
But most of the other booths look downright scary for them. They finished third in most of them, and fourth in at least one.
The model continues to perform very well indeed (predicting final 48.4% ALP 2PP,
standard error of 1.39%)).
The ALP 2PP predicted values for the remaining normal booths are as follows:
New Town 41.78
Rosetta 55.78
St Peters 40.17
Sandfly (Denison) 42.71
Sandy Bay 31.82
Sandy Bay Beach 23.61
Swan 46.89
Taroona 35.51
Tolosa 61.65
Weimea Heights 26.76
Windermere 57.36
New Town 41.78
(3 favour ALP, 8 favour Wilike – the Sandy Bay booths are quite big)
The main hope I can offer labor is that some of the “bad” booths to come have some primary numbers which are outide of the models current “training”. My hope is labor will do better in these booths than predicted which may boost the final ALP 2PP closer to 49, which might give them a shout come postal vote counting..
How did psotals go last time for labor, anyone? Or is that above?
Postals and etc were 65% for Duncan Kerr – very similar to the ordinary vote
My understanding is that in Tasmania (and particularly Denison) postals traditionally favour the Greens. Bob Brown likes to call it the “backpacker factor”. So, Couser might benefit, and when he is distributed, Wilkie (maybe).
Add to this KB’s assertion that Labor largely took the seat for granted, particularly in the matter of garnering postals.
As against this I do know that a sizeable number of Couser voters actually preferenced Jackson over Wilkie (and a lot wouldn’t have).
Any comments?
[Kevin or anyone – would I be right that Tolosa looks good for ALP if they’re still in it after sandy bay count?]
As sykesie’s model predicts, Rosetta, Tolosa and Windermere will all be wins for Jackson.
It has occurred to me that an issue for Wilkie in postals and prepolls will be that the Liberal preferences will not be directed by the How-To-Vote card and may therefore be weaker. I’ll look for comparable data from other elections on this. This could knock as much as 5-6 points off Wilkie’s 2CP from these votes. Not enough to lose him the seat alone but in combination with his likely weaker primary on postals could make things hard for him if his 2CP before prepolls is more like 51 than 52.
It is more important to look at the 1st preference results for
the postals and absents from last time.
http://results.aec.gov.au/13745/Website/HouseDivisionFirstPrefsByVoteType-13745-194.htm
The Greens went ok on the absents but fell down badly on the postals.
Syksie, if you plug 1st prefs for postals from 2007
into your formula, and then the absents, what do they
predict for Wilkie?
Many people do get HTVs from a party with the postal vote
application pack.
I am not sure how to predict pre-polls this time as there
are supposed to be more of them and they are being
counted in with the ordinary votes, I think.
Dr Good, but this doesn’t take into account postals received directly from the AEC (rather than via a party) does it? Of the ‘direct’ postals how many people will have seen a Liberal how to vote card?
Itep
I agree. That is why I only said “many people”.
I think that Laocoon has posted somewhere the
stats about how many postal vote applications
come via party packs. Must be on the AEC
somewhere
Dr Good, based on the values given for postals in the link provided, my model actually predicts an ALP 2PP of 64.5% – which is very nearly identical to what was achieved last time.
This means if labor can go near to maintaining their primary vote of 51% on postals as in 2007, they could overtake a Wilkie lead of approx 1000 votes. Whether labor can maintain a 51% primary must be doubtful, however. The ALP 2PP tanks quickly as the ALP primary drops (the coefficient for that term is 1.136).
[This could knock as much as 5-6 points off Wilkie’s 2CP from these votes. ]
Probably not that much, maybe more like 2. I just looked at Melbourne 2007 where the Liberals, who were the bulk of all preferences, preferenced the Greens in the cutup between Bandt and Tanner. On ordinary votes Bandt got 83% of all preferences and on postals this dropped to 75%.
[Syksie, if you plug 1st prefs for postals from 2007
into your formula, and then the absents, what do they
predict for Wilkie?]
That would not be relevant because the first prefs from 2007 will not be repeated because Wilkie will at least poll *something* on postals, dragging the first prefs for Labor, Green and Liberal down. Plugging the 2007 values into the model assumes a Wilkie vote of zero on postals which is not realistic.
Thanks very much Sykesie. That is encouraging for the
ALP. What about absent votes and pre-polls?
(Provisionals are not important)
Yes. Sorry. Forgot about that.
We first need to apply the kind of average swings seen so far
in booths away from
the 2007 parties and to Wilkie.
[I am not sure how to predict pre-polls this time as there
are supposed to be more of them and they are being
counted in with the ordinary votes, I think.]
I do not think there are that many more than normal in Denison. Most of the prepolls have already been counted and are included in the current primary tally; they appear as the booths Hobart DENISON PPVC and similar. My understanding is that prepolls cast out of division (except for at the Rosny PPVC) remain to be added in. Note that Hobart is the largest of the PPVCs and as expected is strong for Wilkie.
Dr Good – I think the main thing is the labor primary vote. Based on my model, Labor needs to have a primary vote of around 40% to be gaining votes from Wilkie. The more above 40% it is the more votes that are gained. If it drops below about 38/39%, Wilkie seems to hit the lead. This varys a bit depending on the green vote, but is a rough rule of thumb.