New South Wales election: late counting

A post progressively following the late counting in the New South Wales state election.

Thursday, March 28

I’ve been getting my head together the past few days, and have just checked the count for the first time in a while, specifically for the Legislative Council, where the votes are now being counted properly after the election night “initial count” that only gave us a clear sense of above-the-line votes for seven parties. With about 9% of the vote counted on what’s called the “progressive check count”, you would think we were looking at seven each for the Coalition and Labor, two for the Greens and one apiece for One Nation, Shooters, Keep Sydney Open and the Liberal Democrats, with the last seat a toss-up between Animal Justice, the Christian Democrats and Sustainable Australia. However, Kevin Bonham is doing a far better job than I am of keeping score, particularly in adjusting for the fact that the counts in some electorates are more advanced than others. His projections seem to suggest that the Coalition are going to take an eighth seat and that Labor is only assured of sixth; that Animal Justice, the Christian Democrats and Sustainable Australia are all unlikely to win a seat; and that the last three seats are a game of musical chairs between Labor’s #7, One Nation’s #2, Keep Sydney Open and the Liberal Democrats (i.e. David Leyonhjelm).

Tuesday, March 26

Whatever doubt remained about Dubbo, and by extension the government’s majority, was laid to rest yesterday by the 12,319 votes at the pre-poll voting centre in Dubbo, which broke 5493-4728 for Nationals candidate Dugald Saunders over independent Mathew Dickerson. With a further 4581 votes from smaller pre-polls, iVotes and postals breaking almost perfectly evenly, Saunders’ lead has increased from 186 to 960.

Sunday, March 24

I don’t have much of substance to offer yet, but this post will feature regular updates over the next week or two to follow the late counting in New South Wales. The Greens conceded defeat in Lismore yesterday, which surprised me a little, but obviously they would know things I don’t. Clearly then they will drop out at the last count and leave the final count as a race between Labor’s Janelle Saffin and Nationals candidate Austin Curtin (the Nationals incumbent, Thomas George, is retiring). On the notional two-candidate preferred count between the two, Saffin has a surely decisive lead of 3.9%.

Another seat where a Nationals incumbent is retiring, Dubbo, remains in doubt between Nationals candidate Dugald Saunders and independent Mathew Dickerson, with Saunders holding a 0.6% lead. Past history strongly indicates the independent will lose ground on late counting, so this seems very like to remain with the Nationals, but I will continue following the count. A Nationals win would ensure a majority for the government, assuming it stays ahead in East Hills, as it almost certainly will. I will also continue following Coogee, although Labor looks to have that in the bag, one of its only two gains.

Then there is the Legislative Council, to which I currently have nothing to add beyond what I said yesterday. Kevin Bonham has a good account here; no doubt there are others.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

66 comments on “New South Wales election: late counting”

Comments Page 2 of 2
1 2
  1. Lib Dems – KSO – Labor seems to be the popular trifecta in the voting on my Not-A-Poll and could happen but I have some doubts. KSO has dropped back very rapidly but it’s hard to say to what extent this is down to an early vote type skew (iVotes and Sydney Town Hall) washing out and to what extent it is down to divisions they didn’t contest (eg outer suburban ALP seats) being overrepresented in the check count.

    Labor is dropping back in my projection because their BTL vote is falling. Still I would expect that AJP and SUS preferences will go to them much more than CDP and PHON; indeed CDP has nowhere to get preferences from except Aus Conservatives who don’t have enough votes.

    I think the biggest danger to this scenario is KSO crashes out leaving PHON and CDP to fight for the seat they would have won. (At present, I would expect PHON to win.)

  2. Michael, if this were a federal election I would agree with you but I don’t view this election as entirely left vs right, it’s also “liberals” vs the nanny state. I question how much of the past O’Farrell-Baird-Berejiklian agenda Leyonhjelm would have supported and actually look forward to seeing if he and Tyson Koh can form an unholy alliance on several key issues.

    Thanks Kevin, if KSO winning a seat is still the weakest link in that sequence then that’s a good sign for Labor’s prospects.

  3. I agree. I have no idea how Leyonhjelm is going to play it. But I reckon he will still support a lot of issues which suit the right as he did in the senate. But the Upper House is going to be difficult for the Coalition. Andrews in Victoria needs only 3 votes from left wing parties which number 3 or more so and really has no problems getting things through.

  4. If the odd bone for the right is the cost of ending the wars on local councils, greyhound racing, recreational fishing, young people, live music, cyclists, festivals and drugs (amongst other things) then so be it. The upper house can’t do much about all the privatisation anyway afaik, which is the main left vs right issue in NSW imo.

  5. would not be sure the ld win If some one from the “left” won on Michael’s sums 21 all
    but what will the shooters do?

  6. what happens with byelections? and what happens with the next Boundary change?
    any one know the 2pp for Wollondilly? alp not in the final 2
    Hannan and lib

  7. Thanks Kevin, looks like it was an interesting day in the count, but not a huge movement in the projections.

  8. When are the preference distributions expected?

    I’m curious to see what happens in Coffs. While the Nats are certain to get up, it will be interesting to see if Labor, who polled 2nd by only 35 votes, stay ahead of the Independent Sally Townley, who as an ex-Green may well pick up the lion’s share of AJP and Green preferences. If Townley ends up in the last two, it would be interesting to see if her TCP is better than Labor’s 39%. I suspect it would be.

    And it would be good to know whether Lismore’s actually gone as expected to Labor.

  9. Nick Casmirri has tweeted that Green scrutineers in Lismore found that the independent’s prefs were breaking more to Labor than them, which means the Greens can’t get into second and Labor wins. There’s been some talk about postals etc supposedly not being in the count but all that is really going on there seems to be that some of the dec vote forms are down a lot, with prepolls greatly up.

    The scheduled date for Lower House button presses is tomorrow. Upper House is end of next week.

  10. Fred Nile’s bunch are currently sitting on 2.4% (0.5232) vs Kevin’s projection of 2.17 (0.477). Conversely, KSO are down to 2.4% (0.5228) but now projected to actually finish higher on 2.41% (0.530). If not due to error, that’s quite the reversal of disclaimers!

  11. Those are not projected finishing scores. They are adjusted estimates of where parties currently stand – ie where I think they would finish if the rest of the count was the same as what we’ve had so far. We know that because of the stratification in the count that things are almost certainly not going to play out like that. KSO will be lucky to finish above 2% in the primary count; indeed Ross Leedham who has been trying to work out a final position projection has them probably falling behind AJP.

  12. Isn’t that what they’re adjusted for, biases in the sample of votes counted so far? If the projection assumes the rest of the count will be the same as we’ve had so far, wouldn’t the percentages and quota counts be identical to the current raw progress result?

  13. It’s rather complex because there are two counts.

    My estimates are estimates of what the more complete initial count would show now if corrected for the omission of the Others parties, the non-crediting of BTLs to the seven initial count parties, and the non-crediting of non-blank informals to the informal pile. These adjustments are just for the deficiencies of the published data, not for anything that might happen later.

    By implication, my estimates do project that the less complete check count will later change proportional to any change in the share of Others votes needed to match the initial count’s share.

    But they don’t attempt to project any changes in the share of Others votes that will result from some of the Others parties getting a higher share later on in the check count than others. So as concerns the check count they adjust only for whether Others seem to have too many or too few votes collectively, and not whether some Others parties appear to be set to rise or fall relative to the remaining Others parties.

    The reason I didn’t do projections of that kind is that they’re extremely difficult because of incomplete information caused by some of the parties (KSO especially) only running in some Lower House seats, and with no track record of their performance to rely on. Ross Leedham has been trying such projections and doing pretty well with them so far.

Comments Page 2 of 2
1 2

Leave a Reply

Your email address will not be published. Required fields are marked *