I will use this post to provide ongoing commentary of late counting in doubtful seats over the coming days/weeks.
UPDATE (27/3/07): Christian Kerr points to a slow count in today’s Crikey:
The ever-protracted count for the NSW Legislative Council election is likely to be even slower this time, with the Australian Electoral Commission refusing to authorise any more overtime for the AEC staff engaged for the count. There have been unprecedented levels of cooperation between the AEC and the NSW electoral authorities this election, but after just two days of preparing for the Upper House count over the weekend, the AEC has gone into a panic about the likely level of overtime, and has literally ordered its workers to take a “rest”. Counting will now finish each day at 5pm, with no approval for overtime to complete the count. With Easter imminent, this delay is likely to push back the announcement of the Upper House results substantially. The NSW Electoral Commission is understood to have expected the AEC to finish the Legislative Council count by Wednesday. The AEC told staff that the Electoral Commissioner has been informed that he will have to adjust his timetable. No amended timeframe for the conclusion of the count was suggested. A major outcry from furious Government, opposition and minor parties about the delay in finalising the count for the Upper House count in 2003, marred by slow counting and a total meltdown in the computer software used for calculating the results, saw new procedures adopted for the 2007 election. Efficiency was supposed to have been increased by the use of AEC staff in the count.
Legislative Council
| Roy Smith (Shooters) | 83,320 | 0.61 |
| Trevor Khan (Nationals) | 57,727 | 0.43 |
| Arthur Chesterfield-Evans (Democrats) | 50,335 | 0.37 |
| Janey Woodger (AAFI) | 46,332 | 0.34 |
| Robert Smith (Fishing) | 45,460 | 0.34 |
Sunday 3pm. I’m not doing too well here – I now realise the Legislative Council Summary figures I was just getting excited about have been little changed in the past week. They tell us of 3.3 million votes out of roughly 4 million in total, including 293,240 "other" votes that include (I believe) both informals and below-the-lines. The progressive totals figures show us the destination of 13,566 out of a probable total of about 80,000 below-the-line votes; from these the Democrats have polled 5.6 per cent and the Coalition 17.2 per cent, bearing in mind that not all of these votes will stay within the party ticket. Using these figures to extrapolate the as-yet-uncounted votes, I have the Democrats with a fractional lead over the Nationals’ Trevor Khan, but the margin is far too close (and the method far too crude) for anything to be stated with confidence.
Saturday 11pm. Okay, turns out all that effort on the previous entry was wasted. Because as well as the daily PDF file update, the NSWEC also has on its main page a different count with 3,278,467 votes. This includes 293,240 "other" votes, which probably means about 200,000 informals plus yet-to-be-counted below-the-line votes. There would be about 700,000 further to come. These figures show that the Shooters Party are home, while the gap between the Coalition and the Democrats has narrowed considerably. If the Coalition’s share continues to decline at the same rate as it did between the 1.9 million count and the 3.3 million count, the outcome will be very close indeed.
Saturday 10pm. A further 765,023 votes have been added, bringing the total to 1,938,396 out of a likely 4 million. This has resulted in a significant shift in the aggregate vote from the Coalition (down from 35.4 per cent to 34.4 per cent) to Labor (up from 40.4 per cent to 41.4 per cent). If there was reason to think that trend would continue, Labor’s number 10 candidate Barry Calvert might still be out of the hunt. However, aggregate lower house figures (Labor 39.0 per cent, Coalition 37.0 per cent) suggest that won’t be the case, even when taking into account the Coalition’s traditionally lower vote in the upper house (33.0 per cent against 35.0 per cent in 2003). In the meantime, the drop in the Coalition vote has reduced their surplus over the seventh quota from 0.78 to 0.56, almost enough to return the Nationals’ Trevor Khan to twenty-first place, with the Shooters Party up from 0.53 to 0.55.
Friday 8pm. The NSWEC has published a group and candidate votes report, based on the results of 1,168,246 group votes and 5,127 below-the-lines. The totals in 2003 were 3,721,457 and a bit over 70,000. Ben Raue says the two combined suggest the Nationals’ Trevor Khan has moved up a spot from 20 to 21; if this continues, the final spot looms as a race between the Shooters Party (0.53 quotas), Unity (0.35), the Democrats (0.35) and AAFI (0.30), with the Fishing Party slowly but surely headed for the exit (don’t let the door hit your arse on the way out, Bob Smith).
Friday 3pm. Props to Upperhouse.info for pointing out the following message from the NSWEC: "Legislative Council progressive totals will be provided daily in this directory from the evening of Friday 30 March 2007".
Sunday 5pm. The raw numbers at present look straightforward enough: Labor 9, Coalition 8, Greens 2, CDP 1, Shooters Party 1. However, Stephen L in comments cautiously offers that the Democrats (and perhaps also AAFI and the Fishing Party) might do well enough on below-the-lines and preferences to stay in the hunt against the Nationals’ Trevor Khan, eighth Coalition candidate and Poll Bludger fan.
| Greg Piper | 12,913 | 30.3 | 18,656 | 50.1 | ||||
| Jeff Hunter | 17,294 | 40.6 | 18,550 | 49.9 |
Wednesday 2am. One more change of lead in the final strait has given Greg Piper a 106-vote win after the full distribution of preferences.
Monday 2.30pm. Another 940 absent votes have produced yet another change of lead, Jeff Hunter now ahead by 65 votes. Antony Green notes in comments that the closest outcome in modern times was the Liberals’ eight-vote win in Coogee in 1973; this was overturned on a legal challenge, and Labor won the ensuing by-election by 54 votes.
Monday 1.30pm. The lead changes again after the addition of 496 further absent votes, which have put Greg Piper 44 votes in front.
Friday 5pm. In an exciting late-count development, Greg Piper has done very poorly from the addition of 1,988 absent votes (23.7 per cent compared with 30.7 per cent overall), which have turned Labor incumbent Jeff Hunter’s 272-vote deficit into a 22-vote lead.
Thursday 10pm. More than 3000 postal votes and about 700 further pre-polls added; still no absent votes. Greg Piper’s lead has changed little, from 263 to 272.
Wednesday 9pm. Excellent account of today’s slow progress from Sally McEwan in comments, along with informed talk of deep Labor pessimism.
Tuesday 4.30pm. Very good call yesterday from Sally McEwan – the second batch of pre-polls has been very similar to the first, barring a slightly higher primary vote for the Liberals. This boosted Greg Piper’s lead by 243 votes; however, 122 "Dec Inst" votes have reeled him in slightly, going 59-15 in Labor’s favour. Piper’s lead is now 263, but with well over 5000 postal and absent votes pending, it’s still too close to call.
Monday 11.30pm. Sally McEwan corrects my previous description of Carey Bay as a conservative area: "Carey Bay pre-poll is different from Carey Bay conservative lakeside waterfront booth … The remainder of the pre-poll votes will favour Piper in the same proportion or greater".
Monday 10pm. Partial pre-poll results have been posted, 999 votes out of what scrutineer Sally McEwan says is about 2000. These votes are "a mix of Cooranbong and Carey Bay", which is to say they include the much touted Seventh Day Adventist community, along with another conservative area. As expected, these votes have strongly favoured Greg Piper, whose 158-vote deficit has turned into a lead of 64. This sounds a little disappointing from Piper’s perspective, because the remainder of the pre-polls will presumably be strong for Labor. Next comes about 3000 absent votes and 2250 postals – these differed only slightly from the polling booth results in 2003, though Labor’s vote was notably a little lower and the "others" a little higher.
Monday 2.30pm. Looks like those Dora Creek votes for Piper stayed missing – his tally there has gone from 533 to 508. No word yet on pre-polls.
Monday 4am. A scrutineer at the count, Sally McEwan, says in comments she can "confirm the expected advantage to Independent Piper from the pre-poll votes from Cooranbong". These votes "will be counted and distributed tomorrow". McEwan also reports that "24 or so Piper votes" from the Dora Creek booth are "missing", "leading to extra State Electoral officers being called from Sydney for a reconstruction of the Dora Creek booth tomorrow".
Sunday 5pm. Labor incumbent Jeff Hunter leads independent Greg Piper by 158 votes. That would normally be difficult to close, given Labor’s organisational efficiency with respect to pre-poll and postal voting. However, Lake Macquarie has the quirk of the Seventh Day Adventist community at Cooranbong, which produces a big flow of mostly conservative pre-poll votes due to its observation of the Sabbath on Saturday. In 2003, Labor polled 795 votes (34.2 per cent) to the Liberals’ 1173 (52.4 per cent) on pre-polls, compared with overall totals of 54.9 per cent and 30.7 per cent. Pre-polls accounted for 5.1 per cent of the total vote; also still to come are the less quirky absent (7.3 per cent) and postal (5.3 per cent) votes. The latter might go a little better for Labor than last time, as consciousness of their danger might have resulted in a better organised postal vote campaign.
| Craig Baumann | 17,894 | 42.5 | 19,375 | 50.1 | ||||
| Jim Arneman | 17,544 | 41.7 | 19,311 | 49.9 |
Wednesday 2am. The margin widened to 64 votes after completion of the full preference distribution.
Friday 3pm. The notional preference count has been completed, and it points to a 19-vote Liberal victory. However, a "proper" preference count will now follow, and these can turn up anomalies. For example, the primary vote recount cut Chris Baumann’s vote by five votes and Jim Arneman’s by six (UPDATE: And more pertinently, as Geoff Lambert points out in comments, there were variations of up to five votes at individual booths).
Thursday 10pm. Absent and postal votes are now coming in at a fair clip, and while it’s still extremely close, the trend has been with the Liberals. Antony Green’s regular updates show how Labor candidate Jim Arneman’s lead narrowed and then disappeared in late afternoon counting, with the Liberals’ Chris Baumann currently ahead by 56 votes.
Tuesday 8pm. Not much progress today: polling booth re-check completed and 213 "Dec Inst" votes added, increasing the Labor lead from 76 to 86.
Monday 10pm. Either Port Stephens has had an extraordinarily high number of section votes, or the pre-polls have been entered on the wrong line – I will assume the latter. There are 1,244 of them and they have tipped the see-saw back towards the Liberals, whose deficit has narrowed from 153 votes to 76. However, the 2003 figures suggest Labor should do better on absent and postal votes. Slow progress on the polling booth re-check for some reason.
Monday 4am. The Daily Telegraph reports confident noises from a Liberal scrutineer, as "many votes were exhausting because of a decision by the Greens not to preference Labor". Conversely, the Australian Financial Review reports that "Labor strategists are sounding increasingly confident".
Sunday 5pm. Labor’s Jim Arneman was 153 votes behind the Liberals’ Chris Baumann at the close of counting last night, but is now 111 votes ahead. Pre-poll and postal figures from 2003 are probably no guide, as the seat was less fiercely contested last time.
| Jodi McKay | 12,951 | 31.2 | 13,793 | 50.7 | ||||
| John Tate | 10,003 | 24.1 | 13,430 | 49.3 | ||||
| Bryce Gaudry | 8,774 | 21.1 |
Friday 9.30pm. Those two-candidate figures quoted in the Herald have now been posted on the NSWEC site.
Thursday 10pm. Yesterday, the Newcastle Herald told us that "an Electoral Commission notional distribution showed Ms McKay on 13,793 votes and Cr Tate on 13,430". Today it reported that "preliminary counts show that Cr Tate would gain more than 2000 votes on McKay once preferences are distributed". On present indications, that would leave him about 700 votes in arrears.
Tuesday 2am. The NSWEC reveals nothing of the two-candidate preferred count that has evidently been conducted between Jodi McKay and John Tate, but the Sydney Morning Herald reports Tate conceding he is 700 votes behind. Morris Iemma is claiming victory.
Monday 4am. Yesterday’s recheck of first preferences from polling booths has increased Tate’s tally by 18 and reduced McKay’s by 12. The aforementioned Anthony Llewellyn says: "having reviewed the results in total now, my guess is a McKay win over Tate by around 500 … Gaudry will not pull ahead of Tate (of this I am now very confident)". The Sydney Morning Herald reports Labor "has become more confident".
Sunday 5pm. Still anybody’s guess as far as I can see. There is a 2.6 per cent gap between John Tate (24.1 per cent) and Bryce Gaudry (21.5 per cent), which might be closed with preferences from the Greens (11.2 per cent), who directed to Gaudry. Last night’s NSWEC notional preference count assumed Gaudry rather than Tate would finish second; if that is so, Labor’s Jodi McKay will win quite comfortably. If not, the race between McKay and Tate will come down to unpredictable preference flows. Last night, reader Anthony Llewellyn provided a preference breakdown from a booth at which he was scrutineering: if this is applied consistently, Tate emerges ahead with 12,792 votes to 12,327 (not counting preferences from the CDP and three other independents, who collectively account for 915 votes). However, Llewellyn also spoke of better preference flows for Labor at other less conservative booths.
| Pru Goward | 16,994 | 39.9 | 18,632 | 51.3 | ||||
| Paul Stephenson | 10,544 | 25.3 | 17,657 | 48.7 |
Thursday 8pm. Paul Stephenson has conceded defeat after being buried by absent and postal votes, widening the lead to 975. This entry, and the figures above, will not be further updated.
Tuesday 2pm. A further 670 pre-polls have gone rather better for Goward than the previous two batches, increasing her lead by 10 votes. Even better for her are the 154 "Dec Inst Votes" (declaration and/or institution?), which have run 70-31 in her favour.
Monday 10pm. I was mistaken to say all the pre-polls were in – it was in fact only about half. The newly added second batch was not quite as bad for Goward as the first, but it still cost her another 40 votes or so.
Monday 2.30pm. Pre-polls are in (all of them, or almost all), and they are surprisingly poor for Goward – she has polled 35.7 per cent compared with her 39.8 per cent of ordinary votes, while Paul Stephenson has 30.6 per cent compared with 25.1 per cent. If preferences follow the same pattern, this will narrow the gap by 134 votes to a little over 300. In 2003, pre-polls were 5.6 per cent of the total – still to come are absents (8.8 per cent), postals (5.6 per cent) and a few others (0.7 per cent).
Monday 4am. Yesterday’s recheck of first preferences from polling booths appears to have unearthed 38 extra votes for Stephenson and only one for Goward. It appears that Goward is better placed than it seemed on election night due to an across-the-board increase in "plumped" voting (numbering one box and then exhausting) at this election.
Sunday 5pm. An updated count (polling booths only) has seen Pru Goward’s lead after preferences increase from 311 votes last night to a fairly handy 455. Talk of the Labor candidate beating Paul Stephenson into second place on preferences has faded.
| Frank Terenzini | 14,819 | 39.7 | 16,741 | 50.9 | |||
| Peter Blackmore | 10,093 | 27.1 | 16,157 | 49.1 |
Friday 9.30pm. The NSWEC has finally unveiled its notional Labor-versus-independent two-candidate preferred, which shows Frank Terenzini a comfortable 584 votes ahead. That wraps it up for my coverage of this seat.
Thursday 10pm. This count has stayed on ice for some reason, at least as far as the NSWEC website is concerned, but the ABC reports Labor is more than 1,000 votes ahead.
Tuesday 2pm. Very slow progress in the count, but Morris Iemma has claimed victory for Labor.
Monday 4am. The Sydney Morning Herald reports Labor "has become more confident".
Sunday 5pm. As with Newcastle, this is one that will depend on preference flows we don’t know about yet because the notional count was Labor-versus-Liberal, rather than Labor-versus-Peter Blackmore. For what it’s worth, the primary vote figures (Blackmore 27.1 per cent, Labor 39.8 per cent, Liberal 20.1 per cent) are similar to those Pru Goward faces in Goulburn (Paul Stephenson 25.0 per cent, Liberal 39.9 per cent, Labor 22.4 per cent). The difference being that Blackmore will need a strong flow of preferences from the Liberals, while Stephenson will need them from Labor. Can anyone suggest if supporters of one party or the other are more dutiful with respect to how-to-vote instructions?
| Dawn Fardell | 17,158 | 41.9 | 19,270 | 50.9 | ||
| Greg Matthews | 17,518 | 42.8 | 18,590 | 49.1 |
Wednesday 8pm. With most postals and about 600 absent votes now in, any remaining doubt is now gone. Fardell’s lead has now widened to 680 votes, or 0.9 per cent. No further updates will be added to this entry.
Tuesday 4.30pm. Pre-poll figures are now up at the NSWEC site, and they tell a different story to the Financial Review – 2318 for Dawn Fardell and 2177 for the Nationals, widening Fardell’s lead to a surely unassailable 521.
Tuesday 2am. It falls to the Australian Financial Review to inform us that "two-thirds of the pre-poll votes have been counted, according to the returning officer. The results have favoured Nationals challenger Greg Matthews, who garnered 1495 of the pre-poll votes on offer while 1453 went to incumbent independent Dawn Fardell". These results are yet to appear on the NSWEC site. However, this makes only a modest dent in what had been a 401-vote lead.
Monday 2.30pm. Re-checking of polling booth first preferences has now been completed, giving a 42-vote boost to Dawn Fardell. Most notably, 37 votes have been deducted from the Nationals at the Forbes booth.
Sunday 5pm. Independent candidate Dawn Fardell leads Nationals candidate Greg Matthews by 401 votes. The precedent of 2003, when then-independent member Tony McGrane did somewhat less well on non-ordinary than polling booth votes (from a near identical vote total to Fardell’s), suggests this could yet narrow.
542 Comments
Pages: « 1 … 4 5 [6] 7 8 … 11 » Show All
Gonginalong, if you mean what part of the electorate, the answer is “all of it and none of it” – these were absent votes, i.e. those cast on polling day outside the electorate.
Who was this independent in Charlestown? The Greens don’t automatically preference independents ahead of the ALP.
I’ve started crunching the upper results, try and put some quotas up soon.
Hey all,
Got these figures for the upper house. These are simply found by adding BTL votes with ATL votes. I’m assuming that the very small number of BTL votes means that not a significant number will flow away from the ticket they’ve been cast for.
Group A 0.13
The Fishing Party 0.29
AAFI 0.30
Group D (SEP) 0.00
Liberal/Nationals 7.78
Group F (Climate Change) 0.09
Democrats 0.35
Group H 0.02
The Greens 1.85
Unity 0.35
Christian Democratic Party 0.90
Restore the Workers’ Rights 0.17
Group M 0.02
The Shooters Party 0.53
Labor 8.87
Horse Riders/ORP 0.10
Socialist Alliance 0.08
Save Our Suburbs 0.07
Human Rights Party 0.08
Most interestingly, Trevor Khan should be up aruond .7 or .8 of a quota.
After 8 Labor, 7 Coalition, and 1 Green are elected, with 5 vacancies, we get this count:
Fred Nile 0.9
9th Labor 0.87
2nd Green 0.85
3rd National 0.78
Shooters 0.53
Unity 0.35
Democrats 0.35
AAFI 0.3
Fishing 0.29
So rather than the Nationals having to fend off the others, it’s the Shooters. I don’t expect any of those parties to be able to overcome that gap of .18 of a quota, but it’s interesting as well that Unity ended up with a third of a quota.
Lake Macquarie: Ever heard of an “absent pre-poll”?
Can you guess how many there are for Lake Macquarie still to be counted?
No and no, Sally. My expertise should not be overestimated – I still don’t know what a “dec inst vote” is (something to do with hospitals and/or old people’s homes, I assume).
TO Sally McEwan Says
Never heard of an absent pre-poll but I guess there are 4,500 ? Looks like the Court of Disputed Returns will have some work to do this year.
Aren’t they people who pre-polled outside of the electorate?
Declared Institution vote is indeed a hospital or a nursing home, taken over a few days prior to polling day. These are done for Lake Macquarie.
There are still a lot of votes to be counted, and yes, as I predicted, Piper would be disadvantaged by the absentee votes, like most new Independents.
If we had not counted the last 250 of those today Piper would still be in the lead. It was almost a decision to go to the weekend papers in front or behind.
It is that fluid, a la Lake Macquarie.
re absentee pre-poll votes: They are allowed – see s,114P and also s.114Q and a114QA of the PARLIAMENTARY ELECTORATES AND ELECTIONS ACT 1912 –
However must admit they should be very rare – given the circumstances in which a person is entitled to vote pre-poll.
Alex luv, you can quote sections of the Act, yet I would suggest you also study sections of the Bible, as interpreted by Adventists.
Rarity is a relative term, depending on which part of the State of NSW you are familiar with, if any.
This Lake is special.
Sally – I have sailed on Lake Macquarie – I know how special it is – but being an agnostic – my reading of the Bible – whether interpreted by Adventists or any other sect has been fairly limited of late.
Alex, politics is all about “the people”.
Before anyone else gets in, a few more punch lines for friend, ally, colleague, associate, and co-co-conspirator of Milton – old Jeffrey:
“Holy water.”
“Water into whine.”
“Fire, Water, Wind and Earth.”
“The best and safest place in the world – for raising children.”
“Pragmatic ALP.”
There are no secrets amongst the people of Lake Macquarie.
Sally – don’t think I don’t I know a lot of things – I was told about alleged circumstances which form the allegations against Mr Orkopoulos about 2 years before any charges were laid. I have been made aware of other allegations – but at the time I was given the information I was confused because of terminology. Anyway all these things remain allegations only. As such I would prefer not to discuss them as I have no proof of any of them.
Good idea, wait and sea.
“Fire, Water, Wind and Earth.â€
The four elements, yes?
The Mountains, The Lake, and the Beaches.
Ben, given that less than a third of the votes have been counted in the upper house reallocation is there any reason to think these are randomly distributed by area, rather than doing some seats first.
If the vote count is not random and favours coalition seats then the fact that Khan has pulled ahead at this stage doesn’t mean much.
Possibly of more signficance is that so far the ratio of informal to below the lines is running 8:1. Since the Democrats, and to a lesser extent AAFI, woulf be counting on good btl votes to keep them in the race if the number of btls is low then that is bad for them. OTOH the Dems have more than a third as many btls as the coalition at this stage, so there is no doubt these are favouring them – they’d better hope the current low figures indicate that btls are being counted more slowly than other categories.
Sorry, it didn’t click for me that this was only a partial result.
I thought it was all being posted in one part, but come to think of it the total was far below the total number of voters.
I’ve just had a look at it again and Stephen is absolutely correct. This PDF only contains 1.2 million votes, whereas the rough count on the night put the total number of votes at around 3.2 million. So any prognostications I made were too early.
I assume we’re gonna have to wait until Monday afternoon for another update.
Of course if we knew where these came from things might be different – for example if the all the booths beginning with the letter A-f had been counted that should be random enough that things won’t change much. However, if its a case of the seats where the lower house result was going smoothly so the ROs had time to get onto the upper house then that may not be a typical selection of seats.
You can see “Legislative Council Results by District” at the NSWEC site – these results suggest we’re getting a random sample. Every district I looked at had about 2500-3000 votes counted, including those with exciting lower house counts like Lake Macquarie and Port Stephens.
I reckon that the Coalition’s upper house prospects have gone up against their rivals for the last spot. The reason for this is that when the quota is increased, it has far more of an affect on a person who is #8 on a ticket than those who are #1 on a ticket. While it may effect preference flows, the relative number of votes between Shooters, AAFI, Democrats and Fishing Party don’t change when the quota changes. But when the quota goes down, Trevor Khan does relatively better, and vice versa.
So if, as Stephen said, informal votes are outnumbering BTL votes 8 to 1, that’s going to help the Coalition.
McKay vs Tate and Terezini vs Blaxkmore figures are now available on the SEO website. McKay is in front by 363 votes but there are no postal or absent votes in the count.
This is off-topic, but I was looking for the detailed results in the 2004 NSW local govt elections (in particular the City of Sydney) which should be on the NSW electoral commission web-site – and while I get to this page: http://www.elections.nsw.gov.au/local_government_elections/local_government_results/local_government_general_election_2004.html the “Detailed results” page doesn’t load up.
Can anyone do better?
Interestingly, that link sends you back to whichever page you’re on when you try to bring it up. This mystery is beyond my meagre powers.
William when I look at the results by district I see about 30,000 counted – presumably the figures the statewide total of 3 million odd comes from. The only figures I can see in the 2500-3000 range are the “other” pile. Can’t see anything that indicates where the million or so in the above and below count comes from. Am I missing something?
No, I was indeed looking at the wrong column. I did realise not long after that 2500-3000 votes per seat was slightly inconsistent with the total of over a million.
SEO site now lists 3.2-odd million votes counted for LC… Who wants to do the honours?
Re:Newcastle.
I am not sure that McKay’s victory is all that assured. The SEO listings under “post election night” show she did not do particularly well on Absent and Postal votes. If the Green preferences exhaust after Gaudry, I think she’s fine but if they then went onto Tate it will be a close contest.
Christ on a bike – I’ve just spent an hour going through their super-duper daily update on PDF file, and now I see this page has the count at 3.3 million!
Updates added, with about four-fifths of the vote now in. The Shooting Party are home; there is the tiniest flicker of an ember of a remote possibility that the Democrats can still run down Trevor Khan of the Nationals to take the last spot.
That page with 3.3 million, very importantly, doesn’t distinguish between BTL and informal votes.
So unless someone could develop a statistical model which uses this sample of BTL votes to distribute the “others” listed on the front page, we’ll just have to wait and see.
As I said, I still think it depends on the number of BTL votes (which aren’t measured properly in the summary) as to whether Trevor Khan is really in trouble or not.
Ben said:
“April 1st, 2007 at 1:28 am
That page with 3.3 million, very importantly, doesn’t distinguish between BTL and informal votes.
So unless someone could develop a statistical model which uses this sample of BTL votes to distribute the “others†listed on the front page, we’ll just have to wait and see.
As I said, I still think it depends on the number of BTL votes (which aren’t measured properly in the summary) as to whether Trevor Khan is really in trouble or not.”
It’s 5.15 am and here I am reading this blog site again!!!!!
I obviously have missed something because before it went a little off track on Friday I was working on the basis that the Coalition had a .05 quota (wafer thin) lead over the Democrats.
It seems to me that’s what we are still talking about.
pBen said:
“April 1st, 2007 at 1:28 am
That page with 3.3 million, very importantly, doesn’t distinguish between BTL and informal votes.
So unless someone could develop a statistical model which uses this sample of BTL votes to distribute the “others†listed on the front page, we’ll just have to wait and see.
As I said, I still think it depends on the number of BTL votes (which aren’t measured properly in the summary) as to whether Trevor Khan is really in trouble or not.”
It’s 5.05 am and hear I am reading theis blog again!!!!!
I obviously have missed something because before it went a little off track on Friday I was working on the basis that the Coaliyion had a .05 quota lead.
Perhaps the only difference (to this uneducated mind) seems to be is that we now know that the “other category” now seems to be composed predominantly composed of informal votes, with much fewer BTL votes than earlier anticipated.
Have I got that wrong?
Trev
Dear oakeshott country
See my previous post about Newcastle. McKay and Gaudry actually did better than Tate on postals versus election night. McKay’s absentee vote rate was even better in the end.
If one looks at the election versus “non-election” night votes you get the following results (all exclude informal as a percentage):
Election night:
McKay 31.14% Tate 24.14% Gaudry 21.56% Green 11.16% Lib 9.34% Others 2.65%
Non-election night votes:
McKay 31.60% Tate 23.85% Gaudry 19.17% Green 11.39% Lib 11.41% Others 2.59%
Overall these are slightly better for McKay, probably neutral or possibly slightly better for Tate (assuming that he benefits from the better Liberal performance and also assuming that those who voted for Liberal flowed on to Tate without the help of the how to vote) and slightly worse for Gaudry (who would have continued to benefit from the Green preferences altough possibly postal and absentee Green voters may have headed to the ALP candidate in greater numbers).
The notional figures on the SEO website reflect the Pre-Poll and Dec-Inst votes. The Pre-Poll votes are the only post-election votes which went much better for Tate. These are the figures that were also quoted in the Newcastle Herald from earlier in the week.
McKay has done significantly better (on primaries) on the remaining outstanding votes – absentees and postals – with Gaudry being let down on absentees and Tate on postals.
(Presumably the absentee voters didn’t vote for Gaudry because they were all the ones who had moved out of the electorate because they couldn’t stand the guy and just hadn’t changed their enrolment quite yet :;)
You can see the full breakdown of all primary votes now on the SEO site for Newcastle under post-election night votes. The SEO did a full Notional Distribution of Preferences between McKay and Gaudry as late as Friday (if I recall correctly) which reflected the full breakdown of votes and showed McKay extending her lead over Gaudry.
I am at a loss to know why they continued to do this when clearly Gaudry had no chance of pulling ahead of Tate. Perhaps once the RO has determined the 2 candidates they are not able to alter this until all votes are in? This would be the only excuse I would offer up.
Antony Green has McKay now about 1300 votes ahead now of Tate on a full distribution of preferences. I don’t believe that he is computer modelling this anymore. He seems to have access to some information that the SEO does not want to release via its own website.
The whole process must be very frustrating for the candidates. I know I am annoyed as it is my electorate and I would like to know officially who has won by now.
As I said Centrebet has already paid off on my wager on Newcastle. That seems to me to be an indictment of the SEO process in this case.
“Presumably the absentee voters didn’t vote for Gaudry because they were all the ones who had moved out of the electorate because they couldn’t stand the guy and just hadn’t changed their enrolment quite yet :;”
Anthony i assume this means we can expect to see a raft of absentee voters at the next election who have moved out cause they can’t stand Jodi?
I haven’t been doing too well following the upper house count so far – I only just noticed that the 3.3 million count has been little changed in the past week, and is thus old news. So Trevor, you are not missing anything – you have indeed consistently been about 0.05 ahead on this count. The trend to the Democrats I detected by comparing these results and these ones is meaningless, because it was done under the incorrect assumption that the latter was a more advanced version of the same count.
The 293,240 “other” votes include informals, and the precedent of 2003 suggests they include about 200,000 informals and 80,000 BTLs. We also have a grand total of 3.3 million votes out of what should be about 4 million, so there should be about 700,000 ATLs still to come. If the remaining ATLs follow the same trend as those already counted, and if the BTLs follow the same trend as those already counted (only 13,566 out of at least 80,000, so that may not be the case) … well, I’ll only say it will still be very close.
William:
I agree that the Upper House count is a bit of a mystery but I am tempted to believe that what we have been doing (on Friday and Saturday) is comparing apples with oranges. My reasoning is this (and of course I could be wrong!!!).
On the night of the election, at each polling booth there was a count done of the LA votes. There was then to be a count of the ATL LC votes (I assume time permitting). This is what we saw on the night.
Either that day, or the following morning the LA and LC votes were moved from the polling booths to central (second) locations in each area. For instance in the North West the votes were moved from Barwon and Tamworth to Armidale (in Northern Tablelands) where the count continued. I assume at these (second) locations the emphasis was on completing the LA counts. If this is correct then at some central (second) counting locations little LC counting may have been done over the next few days.
I was told (before the election) that the votes taken to the central (second) locations would be “merged” so that after this point it would be impossible to identify votes coming from specific polling booths. What I did not ask is whether this process of “merging” would extend to “merging” LC votes from different electorates.
Anyway, I assume that even if some counting of LC votes was done at these central (second) locations, no BTL counting was started at all (and that makes sense, because what would they do with any data collected).
On about Wednesday of last week LC votes from across the state were moved to a central (third) location in Sydney.
On Thursday the LC count commenced again, at this new (third) location. What I don’t know is what votes they started counting at the Sydney location. It is possible they have started with the uncounted votes from locations where close LA votes precluded LC counts….who knows.
Alternatively, they may have gone back to the start and simply randomly selected bundles of votes and started counting ATL and BTL votes.
Alternatively they may be taking the bundled ATL votes, imputing that data, and then taking the BTL votes from the same locations and imputing that data at the same time.
The problem with this third scenario is that there are above the line votes that will have second , third etc prferences expressed. These votes will have to be separatley inputed into the computer so that the distribution of prefernces can proceed at thye end of the count.
It is this final aspect that is interesting.
On this site we have talked about the BTLs but not much has been said about prefernces. I have no idea how many people have put in a second or third prefrence, but I assume it will turn out to be relatively few. Nevertheless, this factor may come in to play as well.
All very interesting….
Perhaps someone elese may have a better idea about the process and let us know.
Trev
Perhaps someone elese may have a better idea about the process and let us know.. Trev
Antony??? Antony has the numbers for 2003, showing how many are usually in each class, as listed below.
1. “Classic” ATL votes
2. Across the Tops, which NSWEC calls RATL (R for Random),
3. BTLs, which NSWEC calls RBTL
Only 1. was counted on election night.
In the NSW Local Council elections, the RATLs and RBTLs were bundled up (in bundles according to the party of the candidate showing the “1″), and sent to the Counting House. The DROs did keep records of how many were in each bundle, and made them available. There never was a web-site with the numbers available, not even the ATLs. The “Classic” were counted and remained at the local office.
At the Counting House, all knowledge of what was going on was obscured for the duration of the count, about 10 days, appearing in one gynormous bunh at the time of declaration. The main task was to enter an “image” of each RATL and RBTL. Many surprises emerged at the end.
At the 2003 LC election, this was the stage that produced the “meltdown” referred to earlier. One problem was that the software was not configured to reject certain types of errors keyed in by the operators (who were entering “0″s for some papers which were formal according to the Saving Provisions, but which had an allowable number of errors.)
What is happening now is not clear. There is a whole new computer system (EML) for many stages of the election this time round. It has not been completely bug-free.
Thanks Geoff, perhaps it is a little clearer, but only a little.
Trev
What exact formula is used to calculate the quota?
Sunday’s count has 2682934 votes in the count.
Using that as a starting point, is it simply dividing that number by 22, which produces 121952, or is it 121,953 (total/22 + 1), or is it something else?
I’ll add up all the figures from today’s count, and if someone can clarify the quota I’ll produce a quota update.
On my count, the figures are these:
Fred Nile (CDP) 0.93
Mike Veitch (Labor) 0.88
John Kaye (The Greens) 0.84
Trevor Khan (Nationals) 0.63
Roy Smith (Shooters) 0.61
Arthur Chesterfield-Evans (Democrats) 0.36
Robert Smith (Fishing) 0.32
Janey Woodger (AAFI) 0.32
I’m not sure how William is getting the figure putting Trevor Khan just ahead of the Democrats. Maybe I’m making a mistake in calculating how the votes will flow through the other 7 coalition candidates ahead of him.
Quota =1+ (Votes / (N+1)), rounded down
Ben, I’m not looking at those daily PDF file results – I’m looking at these ones. I’m simply calculating a quota by dividing the total of counted votes by 22 (i.e. 22+1), and listing the Coalition’s surplus over seven of these quotas (along with the others’ percentage of one of these quotas). Not doing anything sophisticated involving preference flows. I’ve only just noticed that big juicy updated new file you refer to.
William, Ben, et al,
If I am not being a little presumptuous, and if you had the opportunity of appearing before an Upper House Committee to raise issues regading the running of this election, what would your prime issues be?
Trev
Ben, you are including informal votes in your quota calculation – remove them and you get 117,050. Not that it makes that much difference. Hard to see how Trevor can lose from here.
I can’t see the reasoning in lumping in BTLs with informals. And I can’t work out why the summary figures, aren’t updated by the daily PDF. Or in producing the daily results in PDF! What sort of software are they using that can’t export into XML?
And if Geoff’s tale of confusion is correct (and it fits with my memory of the sorry saga) it looks like it is being replayed.
Fair enough Barney
It was a bit of a tongue-in-cheek spurious explanation for Gaudry’s lack of absentee votes. Possibly a better explanation is that all of his supporters stayed in Newcastle to try and help him. He would have also been disadvantaged by these electors going to other booths and being given a HTV absentee form by Labor or Liberal.
I reckon this tactic probably aids the big parties by a few hundred votes in tight marginals. Seems to be the case in Newcastle and Lake MacQuarie and Goulburn. In the Lib / Lab contest of Port Stephens the absentees are much closer.
I think I managed to give away about 20 absentee HTVs in one afternoon at my booth. The voters who took them all seemed quite thankful that we had some information for them.
William, I noticed that error and fixed it before I posted those quota figures.
The problem that always occurs at this stage of the count is that there are several computer systems in use. One is a simple election night reporting system for primary votes and reconciling ballot papers. The second is the data entry system from BTL and ATL preference votes. The returning officers will tally all the single 1’s and send the ballots and a total into the data entry centre. All the other ballots are sent in as a batch for data entry. And slowly all the numbers will come together.
But trying to get these systems to reconcile and published to the web in a manner that makes any sense to anyone is truly a nightmare. Especially when you remember such eccentricities as a ballot filled in above and below the line will count below the line first, and if that is informal, will become an above the line vote. And also when you remember how few people actually want to look at the data.
The pdf document is the output of their data entry system. It has not been re-developed since the last election and does not have xml output. In discussions before the election, it was agreed that xml for this was not a high media priority, pushing instead for a post-election lower house feed to be developed. (There wasn’t one originally.) That feed worked very well and was the best yet achieved by a state electoral authority.
re LC figures: I got the same results as Ben did from tonight, but comparing to 2003 I noted that (from the numbers we have) the informal rate appears to be down. That being the case, there doesn’t seem to be enough BTL’s counted yet, which may shift the final quota calculations for each party (and who is elected in which spot) somewhat, but not enough to change the actual outcome (at present that Dems are getting about 5.5% of BTL’s, compared to 17.38% LNP). Comparing to 2003 again, there should be another 50000 BTL’s to come.
re Trevor’s question on what to say to an UH Committee? Well, I’m not in the position I was in back in 2003, so didn’t get the reports of RO’s not knowing what was in the Act/what to do, bad rulings, poorly trained staff etc, so I’m not really able to yet say how the two elections compare. Things do seem to have improved, though! I’ll think some more on this before commenting (perhaps more formally!).
Pages: « 1 … 4 5 [6] 7 8 … 11 » Show All