NSW by-elections – what comes next?

24

Last night’s counting for the four New South Wales state by-elections probably set some records for the smallest proportion of the total vote to be counted.

There remains a large number of postal votes to be counted, along with quite a few pre-poll votes. In this post I thought I would run through the coming timeline, what I expect is left to be counted and some early trends that we can see about how the vote broke down.

There will be no more counting today. The remaining pre-poll booths that haven’t yet been counted will be counted tomorrow, on Monday 14 February.

They will also need to conduct a new indicative two-candidate-preferred count for Willoughby between Liberal candidate Tim James and independent candidate Larissa Penn, since the election-night count was between Liberal and Greens. This will hopefully give us a better sense of how close that race is – at the moment we’re assuming that James is leading, but not by a much, on the votes counted so far.

Other special votes (including absent votes, enrol-and-vote votes, other declaration votes and telephone votes) will be counted from Wednesday 16 February.

Postal votes won’t be counted for quite a while. The first count of postal votes will commence on Saturday 19 February, with further counts of extra postal votes as they arrive.

So how many votes do we expect to be left to count? I’m re-using this table I posted last night which estimates the number of votes in each category as a proportion of enrolment. I am assuming a 5% decline in turnout compared to 2019.

StatisticBegaMonaroStrathfieldWilloughby
Assumed turnout84.884.184.284.5
Ordinary total18.420.720.723.1
Pre-poll total29.423.116.39.6
Pre-poll reported so far7.219.310.49.6
Others (so far)0.00.00.00.0
Postal (minimum)21.315.326.428.1
Unaccounted15.525.020.823.7
Uncounted pre-poll votes13,0942,1983,2260
Potential postal/other votes21,71723,41026,09328,533
2CP lead1,9092,0731,324NA
    All pre-poll centres in Willoughby have been counted. We have pre-poll figures from two out of three Strathfield booths, but are missing Burwood. We have pre-poll figures from four out of five booths in Monaro, but are missing Jindabyne. There is a bigger pre-poll vote waiting to be counted in Bega. We have results from Bermagui and Narooma, but not Batemans Bay, Bega, Merimbula or Moruya.
    The NSWEC has already reported how many votes were cast at pre-poll, and every ordinary booth has reported. So if we assume a certain turnout, we can thus calculate an estimated postal and other vote.
    We’re expecting at least 25,000 more votes to be counted in every seat, with almost 35,000 more votes to count in Bega.
    The counting of pre-poll votes in Bega tomorrow should make that seat clearer, and the indicative 2CP count in Willoughby should also make that clearer.
    At that point we’ll likely have a pretty good sense of who is the likely winner in each seat, but we’ll be able to confirm once the first batches of postal votes are counted on Saturday. My figures suggest that a majority of postal votes have been returned already in three out of four seats.

It’s a bit early to say for sure about differences in how people voted between these different voting methods or between different areas. I would normally do some maps, but I’m reluctant to do so considering how few votes were actually cast on the day, but William Bowe has maps on his results pages linked to from his liveblog.

But I have tried to match up election day and pre-poll booths with each “sub-area” of an electorate that I defined in my pre-election guides. This works well for the rural seats, but in the urban seats the pre-poll booths have reasonably wide and overlapping catchments.

In general it looks like Labor has done slightly worse on the pre-poll than on election day, as is traditional. Labor has done worse in the two pre-poll booths to report so far in Bega than in the closest election day booths, suggesting their lead could shrink as the remaining booths report. In the three Monaro sub-areas where pre-poll votes have reported, the gap between election day and pre-poll was about 3-5%. Labor is also doing slightly worse in the pre-poll vote in Strathfield, although Burwood will probably be a stronger booth for them.

That’s it for now. I’ll be recording a podcast tomorrow afternoon to come out on Tuesday morning, and I’ll return with more analysis once the vote count has progressed further.

Liked it? Take a second to support the Tally Room on Patreon!

24 COMMENTS

  1. I scrutineered at Bega prepoll from Bermagii voting centre last night and declared that labor had won bega for the first time based on this result. About 30 minutes later, Antony Green declared Bega a labor win. I would say that Green had been fed this result by labor scrutineers as the data had not been put on the NSW electoral website. What a night to celebrate.

  2. Hello Tallyroom,
    New subscriber, 30 year polling day hawk, though, and 3rd year (mature aged) undergrad of Politics and International Relations, as well as an Eden-Monaro/Bega elector since 1994.

    Firstly, the columns need much better definitions, as 84.8 is referring to what; a single number #, a percentage %, or something else? It is extremely unclear as to what this means or is referring too.

    Secondly, from what I hear on the ground, due to knowing people who worked the booths in various electorates, thus counted the First Preference, Second Preference, etc, im not hearing Labor has done anything like as bad historically within the postal/pre-poll (I’m counting them as essentially the same thing, even though idiotically their [the NSW EC] not counting the former until Wednesday, a decision of which I firmly expect to change by Monday morning due to the small polling day turnout) voting due to many people choosing to not endanger themselves at such an event in an ongoing Pandemic. Clearly, time will tell.

    Otherwise, keep up the good work.

  3. One of the basic principles in democracy is accepting the result and conceding defeat and moving on. Even if the notion of losing is hard. I’m sure the Labor candidate for Bega would have had the decency to concede if he lost but twice in a row Fiona Kotvojs hasn’t has the decency to concede defeat after the result had been called.

    Everyone knows Labor has won Bega, the NSW premier congratulated Labor. So why isn’t Fiona Kotvojs conceding like other candidates has the decency to? It’s over. Unless she can amass an unprecedented amount of support from the Pre-polls and Postals. She has lost this election.

    Please the good and sake of democracy concede the Bega By-election Mrs Kotvojs.

  4. Thanks Denn, the paragraph above the table says “as a proportion of enrolment”. I’m not going to be adding a couple dozen percentage symbols that will just clutter the chart. There’s more explanation of how the table works in yesterday’s live blog.

    My understanding is the postal votes will start being counted on Saturday. Perhaps they will do it sooner but there’s quite a bit to do before that. None of those votes have been opened yet so I don’t know how anyone would know the voting trend amongst them. We don’t even know how many there will be yet! You may be right that Labor will do well with those votes but we don’t know that yet.

  5. Going to have to disagree with you, Daniel. I’m no fan of Fiona Kotvojs but concession has nothing to do with democracy. I’m much more concerned in general about the noble concept of democracy being conflated and confused with the specific way in which we hold and observe elections, and the theatre of concession is decidedly in the latter category.

  6. A concession does not affect the result of course that is what the election is for. But maybe a concession is a formal of manners.The postal and pepoll maybe slightly better for the libs but the stage is reached where the maths make it impossible when all the prepoll is counted this point may well occur

  7. I’m with Ben on this. Daniel, there are possibly half the votes to be counted. IMO, the trends by booth shouldn’t be considered strong enough to accurately measure what the election result is, similar to a number of news media outlets jumping the gun at the 2020 US Presidential election, running off the back of raw on-the-day votes and not taking into account postals at the time.

    Approximately 34k votes left to count with a gap of only 2k? There is a long way to go.

    The other three seats look more certain.

  8. Hawkeye, the same could be said for the other three seats. Anthony Green mentioned it is also possible for the Liberals to win Strathfield, but they would need 54-55% 2PP on the postal vote, a complete reversal of the 53-55% Labor 2PP range recorded in prepolls and polling day votes.

    Whilst postal votes do generally favour the Liberals/Coalition, there has been no record of votes changing that much, so it is highly unlikely the result in any of the four seats will change.

  9. 3 to 1 volunteers to voters ration on polling day in Strathfield, is what they were saying on ABC on Saturday night.
    No way Libs can catch that total if extrapolated- would need exhaustion more than preferences to break even.
    Labor mobilised early with the mail outs, I think managing to sync their own pamphlets with that from the NSWEC. At least that’s the way I received it.
    Perrottet was a massive lag on Sakr’s candidacy who performed much better than I had anticipated, for all the talk of Farrelly garnering the %, I think that also came back to the ineffectiveness of Jason as a campaigner, who looks to have broke even and not capitalised on larger state wide swings.
    When he is on the ballot again in a year, it will give a better indicator of his candidacy, especially now that resources, volunteers and media coverage will be significantly decreased in the general election.

    Both Liberal and Labor volunteers were far too aggressive at the polling booths. There were far too many of them as well at 1 booth with only 2 entries and exits. It felt like you were running some gauntlet into the thunderdome. I think both parties should be fined for breaking COVID rules as well, for not keeping a distance and handing out HTVs at the polls.

    At 46% exhaustion for this seat and Bega, over 40% in Monaro and god knows how much in Willoughby, perhaps the NSW public would be mature to have the FPTP discussion or at a Federal level introducing OPV.

    A lot of people keep saying voter exhaustion is wasting votes but why should voters be forced to allocate preferences if they are dissatisfied, particularly with both major parties? Is that not also demonstrating their democratic right?

  10. Agree with your last point/s LJ Davidson, optional preferences are best because you can decide how many preferences you want to use. The Strathfield result shows Bridget Sakr was a strong candidate and she could win in a favourable environment, particularly if Labor wins federally and/or Perrottet sharpens his image in the coming months leading up to the NSW state election.

  11. As a voter I want to be able who to preference and who not to preference. In 2019, I was presented with dreadful Liberal and ALP candidates. Ideally, i would preferenced neither. In the senate we have OPV why not the house?

  12. The greens ajp and similar do not want a lnp govt yet one encourages them to not extend their preferences. Onp is in most cases a defacto first past the post system which leads most time for the party with the most first preferences to win a given seat.The down point of having to number each square is an increase in informal votes.. this can be avoided by basing formality on the identification of voter intent

  13. Mick quinlivan, OPV is not a direct de facto first past the post system, Labor in NSW won 5 contests (3 in 2015, 2 in 2019) when trailing under OPV and Queensland Labor won 9 seats under the same conditions at 2015.

    There are many voters, mostly from the left, who are in a position like redistributed that dislike both major parties equally.

    A better system is to use proportional representation, either on its own or in conjunction with single member electorates such as MMP in New Zealand

  14. Correct Yoh An, OPV with MMP is optimum. Gives voters greater choice and determination and voter intention is crystalised. It also would lead to unicameralism, which I think is the right thing for states to be moving toward.
    You still get the benefit of minority, negotiating parliaments but save on costs to the taxpayer. The smaller parties and independents would still be safeguarded under MMP. But you don’t have the ridiculousness of 6 and 8 year terms, where politicians become embedded and lazy.

  15. I support OPV for the same reason I support voluntary voting. You should vote, but I respect your right to choose not to. You should express preferences, but I respect your right to choose not to.

    “In 2019, I was presented with dreadful Liberal and ALP candidates. Ideally, i would preferenced neither.”

    This only makes sense if you genuinely believe that both candidates are equally bad.

    Deciding whether to preference candidate A above B is a matter of answering the question, “If the election came down to A or B, who would want to see elected?”

    I find Australian Christians, One Nation, and Australia First all abhorrent choices. But if a seat came down to One Nation or Australia First, I would absolutely want my vote to elect the One Nation candidate. If a seat came down to Australian Christians or One Nation, I would absolutely want my vote to elect the Australian Christians candidate. (If you’re on the other side of the political spectrum, you could say the same about the Greens, the Socialist Alliance, and the Socialist Equality Party.)

    That is why I always number every box.

  16. I agree with you Nicholas, when forced to number all candidates I generally pick the ‘least worst’ of the lot – eg between Liberal and One Nation candidates I always put Liberal ahead of One Nation.

  17. Agree with Yoh An that it does come down to the ‘least worst’ because there is no choice. At one election, in a similar situation, the good lady wife and I agreed to each vote a different way so we cancelled each other out. Neither that or the ‘least worst’ is an ideal situation. It also has to be remembered that we have OPV in the Senate now. You only need to vote up to 12.

  18. Take East Hills 2019 libs outpolled Labor approx 42 to 40. Left of centre parties polled 12% but most exhausted and the lib candidate won.This tends to happen on a macro scale as well. Look at the 4 recent by elections. Same the party polling most primary votes won.

  19. I think Tasmania for Legislative Council elections uses semi-optional preferential voting, you need a minimum of 3 preferences for most contests where there are 4 or more candidates. This system will encourage voters to use up more preferences and reduce some exhaustion rates. It would also allow more trailing candidates to win.

  20. But I believe the better solution is to switch from single member electorates to proportional representation, that way those who want to vote for only minor parties can still do so without risk that one of their opponents will win instead. Many voters who vote for the left of centre parties dont want to be represented by Labor, that is why they wish to exhaust their vote.

  21. You An.. agree with you re pr.. but it won’t happen.. hard to believe left of centre parties prefer a lib or national to Labor the voting system of onp allows that.and in the present climate has a bias in favour of the coalition. Also any one who supports the teals.would prefer a Max preference allocation as this enhances their chances.

Comments are closed.