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”.
Ok well if there is roughly a 13% swing away from the ALP
on 1st prefs (2010 compared to 2007), then only on postal votes is the ALP going
to be near 40%. They could be way behind in absents and
pre-polls.
AEC doesn’t appear to have updated any figures since 9:54. Could they be trying to calm things down ?
[AEC doesn’t appear to have updated any figures since 9:54. Could they be trying to calm things down ?]
I doubt it. I suspect they either have a problem updating the website or else are all busy with something else, given that there have been no updates for any seat. Will check the main thread in case anyone knows.
The ABC page is now confused. It says “ABC Predicted Result: ALP ahead” but shows Antony’s projection of 51.1 2PP to Wilkie.
AEC has 47 of 56 seats now counted on 2CP, Wilkie ahead 51.08%.
all ordinary votes now counted (except special hospitals)…Wilkie leads by 1091 with a 2CP of 51.02%.
is this enough leading into non-ordinary votes?!
sorry should be 51.08% for wilkie
Wow. Sykesie take a bow. A model predicting 48.6% Labor and it comes in at 48.9%
The model finished with a prediction of 48.6% 2PP for labor – so it didn’t do too badly!
(not bad for only 2 variables – green and labor primary)
1091 looks to me like it will be enough to keep Wilkie in front. I reckon Jackson is going to need a primary of close to 45% on the remaining ballot papers to win from here.
I really hope I’m wrong.
Good work by sykesie – whose last prediction was for ALP at 48.4%! Within 0.52%. I think it’s now very hard to predict the likely absent vote and postal vote count based on data from the previous federal election, but I wonder whether data from the last state election might be useful? Very conveniently the state uses the same electorates and Wilkie, Greens, Labor & Liberals were all candidates.
BTW, in relation to comments about the quality of MSM reporting: it’s not actually the local knowledge available here which makes the difference, it’s the application of common sense and a little time invested in examination of the data, together with some basic maths and statistical work. Any journalist with basic skills who wanted to work this out could do it themselves. But they don’t invest time because they are busy thundering off in pursuit of the herd lest they miss the latest ‘important’ development, like what the ABC computer says!
My model, to be fair, wasn’t as good as Antony’s, as he pretty much had it spot on at 51.1% for Wilkie. However, for a 2 variable model, I was pretty surprised that mine worked as well as it did.
Sykesie. Great job.
Antony might have been predicting the final result
not just the booth totals. So you still might be closer.
I also reckon Jackson needs primary of 45%
159 spot on
160 spot on also
sykesie, Antony’s projection took a while before it settled onto the right answer – earlier he had it as 50.3 to the ALP. His seemed overly sensitive to individual booth results.
I had been playing with the one-variable version of your model (ie, just using ALP primary% to predict ALP TCP%), and it also worked remarkably well, predicting an ALP TCP of 48.6 after two booths, and never going in favour of the ALP.
skyesie
Amazing job.
It would be very bad for your self esteem to compare yourself to Antony, but you were quicker than him to get to the conclusion that the ALP would get below 49% of the TCP.
Lets hope the ALP can get home on postals.
[is this enough leading into non-ordinary votes?!]
I’d like to see how he goes on prefs compared to his primary on the prepolls.
[sykesie, Antony’s projection took a while before it settled onto the right answer – earlier he had it as 50.3 to the ALP. His seemed overly sensitive to individual booth results. ]
Yes Antony still had Labor ahead quite late last night on projection whereas sykesie’s settled down extremely quickly.
Rob Oakeshott has only been out and about to the greater world for a few days but he is already annoying me. Why don’t we all hold hands and sing songs next Rob?
Bipartisanship was such a ploy for Gillard and Rudd. All garbage on their own terms of course. Let’s get on with it Rob. You either play or you don’t. Forget the running commentary!!
[Yes Antony still had Labor ahead quite late last night on projection]
I do Antony a disservice there. 4pm.
In support of canberra boy’s comment above at #160 :-
I know absolutely nothing about Tasmanian/Denison politics and demographics but by coming here I have been credibly informed about what is really going on.
If I can do it so can any journalist all of whom have more time, resources and personal connections than me and therefore more able to get and diseminate information rather than piffle.
Copy and paste from what I just put on main thread:
[On Denison and pre-polls: We already know the destination of 4437 prepolls that were counted on the day. These split Wilkie 924 Jackson 1558 Barnes 40 Simpkins 1102 Couser 813 and are already factored into the primary totals but not yet into the Jackson vs Wilkie two-candidate preference count. We also already know the destination of 595 hospital votes yet to be added into the 2CP. So these still have to be distributed between Jackson and Wilkie before we have a preference count that matches the known primaries, and before postals.
Beyond them there are potentially 10349 votes to be added as shown here:
http://vtr.aec.gov.au/HouseDivisionDecScrutinyProgress-15508-194.htm
The big concern for Wilkie in my view on all these extra votes is that the Liberal preferences won’t flow as strongly to him in the absence of how-to-vote cards. But when the existing prepolls are added to the 2CP we will have more idea about this and how much of a problem it might be for him.]
And another one!
[Re #3288 I spoke too early, they have updated and the 2CP now includes those prepolls. Wilkie did not suffer and has extended his lead to 1375 votes.
The mountain for Labor is steep. By my count they need 56.6% 2CP over Wilkie of all the remaining votes assuming that total of votes remaining is accurate.]
(and also assuming Wilkie is not KOd at an earlier stage of counting.)
sykesie do you have time to run your model over the pre-poll figures and the special hospital booths? The primary votes have been counted and are available via the AEC booth pages, but they haven’t done the TCP.
I’ve had a look at the March Tasmanian election results for Denison. Primary vote percentages for party groupings below: each line shows firstly polling booth percentage results, then postal votes, then ‘out of division’ (equivalent to absent) for each party.
Wilkie 8.44% / 7.73% / 7.01%
ALP 36.30% / 39.65% / 33.15%
LIB 29.79% / 32.55% / 30.46%
Green 24.89% / 19.62% / 28.64%
SA 0.57% / 0.45% / 0.73%
What this shows is that Labor got a postal vote result that was about 9% better than the booths, and an absent vote result that was 9% poorer, while the Greens did 11% worse on postal votes compared to the booths, and 15% better on absent votes.
There’s some 10,349 currently outstanding postal, absent, provisional and outside electorate pre-poll votes (this number is still rising as information presumably trickles in to Denison from elsewhere). Labor would need over 55% of these to make up the present 1091 vote gap. Hospital & pre-poll data of course will change this.
[There’s some 10,349 currently outstanding postal, absent, provisional and outside electorate pre-poll votes (this number is still rising as information presumably trickles in to Denison from elsewhere). Labor would need over 55% of these to make up the present 1091 vote gap. Hospital & pre-poll data of course will change this.]
The hospitals and the in-electorate prepolls are in the 2PP count now. See #172. There are still apparently 1427 “Early Vote (Pre-Poll)” envelopes outstanding, whatever they are.
@173
(and also assuming Wilkie is not KOd at an earlier stage of counting.)
How likely is this? It looks like Wilkie’s about 1500 primary votes ahead of the Greens. Would they really make up 1500 primary votes on postals and absentees?
[@173
(and also assuming Wilkie is not KOd at an earlier stage of counting.)
How likely is this? It looks like Wilkie’s about 1500 primary votes ahead of the Greens. Would they really make up 1500 primary votes on postals and absentees?]
No they wouldn’t. But the lead is misleadingly large because the Socialist candidate has 745 votes and they will flow strongly to the Greens. So the Greens may only need to make up 1000. Even then, that is a tall order.
There is a concept in Tassie called the “backpacker vote” which suggests that Greens sometimes improve their position dramatically on non-ordinaries as Green postals flow in a week after the poll from East Timor, Brazil, Slovakia or wherever. I think it is a bit overemphasised (often the Greens will flop on postals) and the main argument for it comes from Christine Milne in the Senate in 04, and how she pulled ahead of Family First for the final spot, going from a position of some danger to having several thousand votes to spare. But that was actually mostly generated by changes in the positions of other parties, and may reflect that the Greens took a dive in support during the campaign (after some of the postals had been sent) because Latham’s forests policy took votes away from them.
As for stage two of the count, based on Simpkins’ performance on the known prepolls I expect him to stretch his primary gap by a few to several hundred as the postals come in. But Simpkins performed so poorly on preferences in the notional 2PP vs Jackson, even with neither Wilkie nor the Greens directing, and the evidence is already strong enough that the Green voters prefer Wilkie over Jackson, that I believe Wilkie will blow away any Lib lead less than 1000 votes in the three-way split. Even 2000 could well be catchable though that’s probably a rock Wilkie wouldn’t want to turn over, just in case.
Go Wilkie!!!!
By my reckoning Labour well and truly stuck on 72.
Maybe tonight might be the time for the proverbial to hit the fan. Definitely time for post mortems. Maybe Mark Arbib might wan’t to do a full hour on Q and A with questions from the audience. Interesting viewing.
Bitar might wan’t to finish that course off in the states in political spin and electioneering?
Bill Shorten might actually show some contrition and admit he made a mistake. Pigs may fly too.
Paul Howes may also become an advocate for mens health issues. Or Jenny Craig
the AEC website has awarded 1 seat in Tassie to the Independants… does this mean wilkie is over the line?? it had 4 independants as having ‘won’ seats….:(
[the AEC website has awarded 1 seat in Tassie to the Independants… does this mean wilkie is over the line?? it had 4 independants as having ‘won’ seats….:(]
No, it means the AEC website has wrongly called it for him because he is 51.2% 2CP. It only considers seats close when they are inside 50.5% 2CP. Yesterday it was wrongly calling it for Wilkie.
We still don’t know if Wilkie will stay ahead after postals.
We still don’t know if the Green will catch him on postals and preferences from the Socialist.
We still don’t know if Wilkie will catch the Liberal on Green preferences after prepolls and postals.
My advice to everyone trying to follow the count:
1. Ignore any call of the result by the AEC computer
2. Ignore any call of the result by the ABC computer
3. Ignore any call of the result in the mainstream media
This thread will say when it is clearly over, and until then it is up in the air.
I clearly spent too long on myvlast comment and didn’t catch the pre-poll & hospital figures posted by AEC!
KB (#172) I think it may be steeper than 56.6% because a proportion of the 956 provisional votes will be rejected. These are votes cast by people who turn up at a polling place, find that they are not on the roll, and assert that they should be on the roll (ie that AEC has mistakenly removed them). I can’t readily find figures from the last election, but my memory is that AEC usually finds a high proportion are not eligible.
When we put the Tasmanian election postal and absent figures at #174 above together with Laocoon’s ‘caught napping’ comment (#4) it looks unlikely that Labor could get there.
I agree with KB that the one remaining imponderable is whether Green preferences put Wilkie ahead of the Liberals: on current primary figures he needs to catch up 200 votes in this distribution.
Looking back to the State election (again), I found that preferences from the four Green candidates who were excluded in Denison went 2684 Wilkie and 1412 Liberal (65.5% to 34.5%) – and obviously some to ALP as well. The unfortunate problem is that at the last Green candidate’s distribution (which was the overwhelming bulk of the total), there were no Labor candidates left in the count, so we don’t know how many of the preferences for Wilkie and the Liberals would first have gone to the ALP.
I think we don’t have enough data to be able to predict whether Wilkie will make up the 200 votes or not. Instinct suggests he will.
[Yesterday it was wrongly calling it for Wilkie.]
Meant to say yesterday it was wrongly calling it for Jackson. Today it is wrongly calling it for Wilkie. The correct call is “We. Just. Don’t. Know.”
On postals: I see the Libs have 1286 party postals in Denison compared to just 80 for Labor. (Strangely, the Libs put lots of party postal effort in in Denison and Lyons). Since the Lib voters like Wilkie more than Jackson this is good news for Wilkie in the 2CP contest with Jackson. But it is not such good news for him in the fight for third against Simpkins, as it increases the (slim) chance that the margin between Simpkins and Wilkie on primaries (currently c. 200) could blow out to something four-figure that is actually some sort of challenge for Wilkie to catch on Green preferences in a three-corner contest.
Libs sent out postal vote applications right across the state to try and save the 3rd senate spot. ALP focussed on Bass and Braddon in their postal vote campaign.
Given its contorted, labrynthine nature, maybe Oakeshot should include urgent change of the voting system covering Denison as part of his reform strategy.
[KB (#172) I think it may be steeper than 56.6% because a proportion of the 956 provisional votes will be rejected. ]
Yes and also some postals may not be returned.
[I agree with KB that the one remaining imponderable is whether Green preferences put Wilkie ahead of the Liberals: on current primary figures he needs to catch up 200 votes in this distribution.]
I expect it to be much more than 200 by the time the count starts. But anything less than 1000 will be a piece of cake. Perhaps the gap to close will be more than that though (see above.)
I base my views on how easily Wilkie will catch the Lib (given a modest target) not only on the state election result but also on the Jackson/Simpkins 2CP. I forget exactly where Simpkins’ 2CP finished up on that (it’s been taken down) but there was actually a swing to Jackson, albeit a tiny one. I think the Lib 2CP off a primary of 22 was only about 34, maybe even 33. That means that the preferences of Wilkie, Couser and Barnes combined split something like 70:30 to Jackson over Simpkins even when they can’t be thrown to Wilkie. The Barnes prefs go almost entirely to Jackson in that split but they are a minority, and I’d expect Wilkie’s prefs to be more friendly to the Lib than the Green prefs simply because he is more moderate. So that suggests the Lib is only getting low-to-mid-20-something percent of the Green prefs, if that, in a straight contest with Labor, which makes sense since the Green prefs split 83-17 in Denison last time.
Throw Wilkie into the mix as well and that takes more Green prefs off the Lib and it’s very hard to see how the Libs even get 20% of the Green prefs in that three way split. And we already know the rest don’t go *overwhelmingly* to Jackson since if they did then Wilkie wouldn’t be beating him on 2CP off the combined Lib and Green preferences. (If I get around to it I’ll do a regression and try to get a handle on the actual figure). Most likely they go more to Wilkie than Jackson. In that threeway split Wilkie-Jackson-Simpkins on Green prefs I just can’t see why Wilkie minus Simpkins would be anything less than 15% of the Green total and it could be more like 25% even. Plus the Socialist prefs will favour Wilkie ahead of Simpkins, heavily.
[Looking back to the State election (again), I found that preferences from the four Green candidates who were excluded in Denison went 2684 Wilkie and 1412 Liberal (65.5% to 34.5%) – and obviously some to ALP as well. The unfortunate problem is that at the last Green candidate’s distribution (which was the overwhelming bulk of the total), there were no Labor candidates left in the count, so we don’t know how many of the preferences for Wilkie and the Liberals would first have gone to the ALP.]
That includes leakage out of the Greens ticket at a point where there are other Green candidates remaining, and those are probably more personal votes for specific Greens than Green-ticket votes as such. The split if you just look at the Green votes that left the ticket without exhausting is not much different (slightly better for Wilkie). Wilkie’s profile was lower then, but it’s possible his 2CP vs Archer would have been even better if Green voters weren’t allowed to exhaust their ballots.
[Given its contorted, labrynthine nature, maybe Oakeshot should include urgent change of the voting system covering Denison as part of his reform strategy.]
It’s the same as the voting system everywhere else. We just happen to be the only seat in the nation that’s thrown up such a complex scenario this time. And if it’s ever happened here before it would have been a long long time ago.
Denison can be proud! This is the second election this year we’ve led the way in producing an obscure cutup. This one is even messier than the state election and that was using Hare-Clark!
[And if it’s ever happened here before]
In a federal lower house election, I mean.
[Libs sent out postal vote applications right across the state to try and save the 3rd senate spot. ALP focussed on Bass and Braddon in their postal vote campaign.]
The AEC figures show the Libs focused on Denison and Lyons with a lesser focus on Bass and negligible in Braddon and Franklin. So I guess they decided the best bang for their buck in Senate terms was to target the seats they knew Labor would put no effort into.
KB (#185)
The sykesie model suggests that the Greens preferences go somewhere around one-third to Jackson in a Wilkie-Jackson split: I realise it doesn’t say exactly that given the two variables in the model, which also accounts for Lib & SA prefs, but it’s close enough for our purposes. That confirms your estimate of Wilkie minus Simpkins being worth 15% to 25% of the Greens total. It’s looking very good for Andrew Wilkie unless he has suffered a total disaster versus the Greens on postals and absents.
And on that subject, the Denison divisional office have now received all of the absent and provisional votes and most of the postal votes, so we should see some more great counting action tomorrow!
I plugged the prepolls of Glenorchy and Hobart into sykesie’s model and the 2PPs for Jackson were as predicted within a few points (one was even lower!)
Unless the prepoll centres had HTV cards being handed out then the prepolls at least provide no evidence that the Lib preferences flowing less strongly because of no HTV card is a big issue for Wilkie. Indeed, in Glenorchy PPVC, his 2CP is over expectations. Which is odd, because it should do him *some* harm, but I can think of a possible counter-factor. That is that Jackson was so low profile early in the campaign.
I’ll be on the road for a few hours in the middle of the day tomorrow, and then staying in Latrobe. Have a wireless modem but not sure what my access will be like. I’ll try to take a peek in Campbell Town at least!
Anyway, unless it’s really really close, counting most of the postals tomorrow should give us an answer to steps 1 and 3 of the process and some kind of indication of what Wilkie has to do for step 2.
KB I would imagine that in Denison the parties would have had pre-poll centres covered with HTV. They always have them here.
[KB I would imagine that in Denison the parties would have had pre-poll centres covered with HTV. They always have them here.]
OK, in that case the postals angle is still interesting.
Ignoring the provisional votes, many of which will be turfed and the rest of which may not do much either way, we have potentially 9663 votes not infected with HTV. However only 7224 of those are currently received. Does anyone know why only 97 of 1618 “Early Vote (Pre-Poll)” envelopes are showing as received?
Let’s assume all those 9663 votes are in and formal, and let’s also assume the standing of the parties on primaries in the postal is the same as for the election overall (which it won’t be, the Libs will probably be higher in particular). Jackson needs a 57:43 split of what’s left compared with 48.8:51.2 normal. So he needs 4.1% of the 2CP vote to flip from Wilkie to him compared with in the other votes. For that to come entirely from the Liberals means that the rate of Liberals preferencing Jackson over Wilkie without the HTV card is 18.6 points lower than the rate of Liberals preferencing Jackson over Wilkie with it.
http://results.aec.gov.au/13745/Website/HouseStateFirstPrefsByPartyByVoteType-13745-NAT.htm
You may want to use these percentages to calculate the number of formal votes for each vote type.
KB (#192) I think the reason for only 97 of 1618 pre-poll votes being shown as ‘received’ is that these were cast outside the electorate. The Denison divisional office is still waiting to receive the bulk of them from the other divisional offices. You’ll recall that the pre-polls cast in the electorate itself were counted on election night and are shown in the booths list as ‘PPV’ or ‘pre-poll’ booths.
OK, got that. I’m losing track of time here and didn’t realise we’d still only had two working days of counting and hence those coming in from prepoll centres interstate probably wouldn’t have arrived yet. Thanks.
I noted Antony has practically called this for Wilkie on TV tonight. I’d like to see some postal results first.
KB the other close ones are interesting, too. Have a look at Brisbane and see what you think. I reckon it will go to Labor. Hasluck not so good. Corangamite *shrug*.
Thanks to people here for the comments, being from Denison but currently in New Zealand, it has been really good way of keeping up with what’s happening. Hopefully Wilkie can get over the line and Southern Tassie can finally get some representation.
Fat lady time on this one
ALP candidate on radio this morning conceding defeat
Report of Jackson’s comments here:
http://news.smh.com.au/breaking-news-national/labor-ready-to-concede-denison-20100825-13qzd.html
Not quite a concession he has definitely lost (unless there is more to it than reported there) but an accurate appraisal of the situation in my view.
Conceding defeat at this stage is slightly strange, I would have thought. I’d be very surprised if the ALP had briefed their scrutineers ahead of Saturday to look at how many Greens votes go to the Libs versus Wilkie. They would have been checking how many went to the ALP. And the ALP would have no idea what’s in the postal and absent votes in the Greens vs Wilkie contest to see who is first eliminated. Either Labor’s organisation in Denison is far more sophisticated than I thought, and has good information, or they have fallen into the trap of basing a decision on the 2CP distribution without questioning whether Wilkie will stay in the real preference distribution to the end. The work reflected in this thread suggests Wilkie will probably win, but there’s still some uncertainty.