Saturday, February 19
The Willoughby count has been updated with 9673 postals on the primary vote, and while these have leaned less heavily to the Liberals than postals did in 2019, they suggest Tim James’ lead is likely to widen from here rather than narrow. James has added 4357 primary votes, or 45.0% compared with 43.4% of all either votes, while Penn has gained 2651, or 27.4% compared with 32.2%.
End of Thursday
Counting has completed in Willoughby for what I expect will be all votes other than postals, leaving Tim James on 8365 (51.76%) to Larissa Penn’s 7797 (48.24%), with 2085 votes having exhausted. The total of 18,247 votes counted is put in the shade by the 25,071 postal votes that have now been received, which will be supplemented by a small number over the next week, counting of which will finally begin on Saturday.
End of Tuesday
I’m bumping this thread back to the top of the page because the NSWEC is conducting a new two-candidate preferred count in Willoughby between the actual leading candidates, Liberal Tim James and independent Larissa Penn, having conducted what proved to be a superfluous Liberal-versus-Greens count on the night. With results reported from 14 out of 24 booths, preferences are flowing heavily enough against the Liberals (10.2% to Liberal, 42.7% to independent and 47.1% exhausted) to perhaps make James nervous. The numbers are available at the NSWEC website – they are not being published to the media feed unfortunately, so my own results display doesn’t show them. James presently leads 51.4-48.6, which increases to 51.9-48.1 if the preference flow so far is projected across the yet-to-report booths.
That leaves, so far, 18,033 postal votes that have been received, which is more than the sum of the votes so far counted with many more still to arrive, none of which will be counted before the weekend. Ordinarily I would point to the fact that Gladys Berejiklian polled 66.5% of postals in 2019 compared with 57.6% of election day and pre-poll votes and deem the door still closed. But that was from a total of a mere 2269, in contrast to the present extraordinary circumstance where postal ballots were sent to every enrolled voter in the electorate. It may be that, as in the United States, that pool of postal voters, which is traditionally older and more conservative, is now dominated by those most concerned about COVID-19.
For more on the by-elections, below is a discussion between myself and Ben Raue which was recorded on Monday afternoon:
End of Saturday
Counting closed last night with quite a bit of the pre-poll vote still unreported – I’m unclear as to whether that will be finished off today or tomorrow. In the very unusual circumstances of these by-elections, there will then be very little to report for a full week, since counting of the unprecedentedly large number of postal votes will not start for another week, as was explained here. There may be a few other types of vote added to the count (absents, provisionals and telephone voting), but these will be few in number.
What we may see is a preference throw between the two leading candidates in Willoughby, Liberal candidate Tim James and independent Larissa Penn, to supersede the redundant count between the Liberals and the Greens that was conducted on the night. James leads by 43.5% to 32.2% on the primary vote, which as it stands might be dangerous for him under full preferential voting, but exhausted votes will result in a weaker flow to Penn. Besides, postal votes are likely to widen the gap, with Gladys Berejiklian having polled 66.5% on postals in 2019 compared with 57.6% on election day and pre-poll votes.
The headline-grabber is Labor’s win in Bega from a thumping swing of 14%, based on all election day votes and two out of six pre-poll booths. Michael Holland duly gains the seat formerly held for the Liberals by Andrew Constance, who will now run in Gilmore at the federal election. It was a better night for the Nationals, who have held Monaro against a 6.3% swing, with four of six pre-poll booths in together with the election day vote. This is a fairly creditable result, given they have held some of the 9.1% swing John Barilaro picked up at the 2019 election.
Labor appears to have held Strathfield with no substantial swing either way, leaving Jason Yat-Sen Li with a 4.4% lead based on all election day votes plus two out of three pre-poll booths. There’s presumably a good reason why no one is countenancing the possibility that the avalanche of outstanding postal votes might overturn this, but it’s very far from mathematically impossible. This would ordinarily be reckoned a soft result for an opposition party at a by-election: mitigating circumstances might include a particularly appealing Liberal candidate, the popularity of outgoing Labor member Jodi McKay, and the fact that independent Elizabeth Farrelly, who was recently a Labor council election candidate, polled nearly 10% and recommended an exhausted vote.
showhide more_text="Show election night commentary"
Election night commentary
10.40pm. A second pre-poll booth from Strathfield, Strathfield EM Office (as distinct from Strathfield EVC), is much more favourable for Labor than the first — a 10.7% swing for compared with 4.8% against — but with less than a quarter as many votes.
10.31pm. Two more pre-polls, Bungendore and Jerrabomberra, are in from Monaro on both primary and two-party — Jerrambomerra was bang on the 7.9% election day swing, Bungendore was 13.3% but with fewer votes.
10.27pm. The Strathfield pre-poll is now in on two-party, and it was a 4.8% swing against Labor, such that there is now a 1.4% swing against them in the seat overall.
10.24pm. Primary votes from the Narooma pre-poll centre, the second in Bega out of six, are much the same in swing terms as the election day results, discouraging the notion of such disparity between them as might jeopardise Labor’s substantial lead.
10.01pm. The last of the Willoughby pre-polls is in on the primary vote, and contrary to what I suggested earlier, the Liberals have suffered identical 15% primary vote swings on election day votes and pre-polls.
9.43pm. The first pre-poll result from Strathfield, primary votes from the Strathfield booth, are quite bad for Labor: down 9.8% on the primary vote compared with 4.3% on election day.
9.32pm. Queanbeyan pre-poll in on two-party: 3.7% swing to Labor compared with 7.9% on election day.
9.24pm. A second pre-poll booth, Queanbeyan, is in on the primary vote in Monaro, recording a swing against the Nationals slightly below the election day results.
8.49pm. The first pre-poll booth from Bega, Bermagui, is in. My results page says it has swung no different from the election day booths, but this is the one pre-poll in the seat that was not in use in 2019 so I’d advise some caution. Nonetheless, Labor candidate Michael Holland is claiming victory as I speak. Regarding the previous entry below, it seems my perception of a weak result for the Liberals in pre-polls in Willoughby is to do with the way I’ve calculated the historic results, which was a necessarily imperfect process. The remaining pre-poll booth there, North Willoughby, will presumably tell a different story.
8.16pm. A second pre-poll booth is in from Willoughby, which makes two out of three, and I’m now more confident that both show a drop in the Liberal Party vote similar to the election day booths, and pre-polls are not in fact transforming the result here. So I remain curious as to where Zimmerman got his estimate of a 61-39 result.
8.07pm. The first pre-poll booth from Willoughby, and the second overall, is Willoughby EM Office – my swing figure says it’s not much different from the election day results, contrary to what we were hearing from Trent Zimmerman. However, my historic figures for pre-poll booths in the city seats are dubious because several new booths were added, and discretion was needed in dividing up the 2019 votes.
7.59pm. All election day votes are in from Strathfield now on both primary and two-party, and with the usual caveat added about election day votes only, the lack of any swing is at least a partial qualification to the picture of a night for triumph for Labor.
7.49pm. My commentary has slowed to a trickle because the picture on election day votes is now clear. We’re now in a period typical of modern election nights where there’s a lull between the election day votes coming in and the much larger pre-poll booths reporting later in the evening.
7.39pm. Trent Zimmerman goes so far as to suggest Tim James will end up on 61% two-party preferred, which suggests an extremely different dynamic on pre-polls than election day votes.
7.36pm. Antony Green in the same position as me in relation to Bega — his system is calling it for Labor, but the peculiarities of this election are such that he’s not willing to shut the door.
7.33pm. The Cooma pre-poll booth from Monaro is in, and there’s only a very slight swing to Labor compared with 7% on the election day booths — the first evidence those of us without scrutineering contacts have had that the Coalition may indeed be doing better on pre-polls.
7.26pm. I’ve spent a few minutes trying to fix my preference flow table, without success, so just disregard that. So far as the election day vote, the picture is now fairly settled. Firstly, the number of election votes we’re talking about here is unprecedentedly small, which has been reflected in the speed of the count. But for what they’re worth, the Liberals look like they’re in all sorts of trouble in Bega, but the ABC coverage indicates the feeling within the party is that this is misleading. There’s been a lot of progress in the Monaro count since I last commented, but no change in the overall picture — a big swing to Labor, but apparently not quite big enough. Very little swing in Strathfield; Liberal Tim James leading independent Larissa Penn in Willoughby by about 42.5% to 32.%, which would leave Penn in need of a very strong flow of preferences, so you wouldn’t rule her out unless you accepted the Liberals’ insistence that they’re doing better on pre-polls.
7.12pm. A slight swing to Labor in Strathfield; things have settled down in Willoughby, where my impression is of a tight result with maybe a slight edge for the Liberals over the independent, with the inevitable qualification that we’re flying blind without pre-polls and postals.
7.11pm. I’ve fixed my projections issue, although the swings in my “preference flows” table aren’t working, which was related to the problem with the projections. Consistent with what’s being said by Trent Zimmerman now, I’ve got the Nationals probably home in Monaro, but Bega still looks very bleak for the Liberals.
7.03pm. There’s an error in my projections which I’ll now work on fixing. Pay attention instead to the results in the tables.
7.01pm. Now looking even worse for the Liberals in Bega, where I think my numbers might be more up to date than the ones just noted by Antony Green, which had the swing at around 8% rather than my 12%.
6.58pm. Better news for the Liberals in Willoughby now. With 11 booths in out of 24, Tim James holds a fairly solid lead on the primary vote, which would probably be enough for him if there’s any rate of preference exhaustion, and certainly would be if it’s indeed the case that he’s doing better on pre-polls. Note that I’ve now arbitrarily set the probability gauge here at 50%. Also no swing now in Strathfield.
6.57pm. I’m now projecting a squeaker in Monaro off five two-party and nine primary results, which you can take as a Nationals win if the pre-poll and primary dynamic is any better for them.
6.56pm. That update is through now. We’ve now got six two-party booth results now from Bega, compared with I think only two before, but the situation hasn’t changed — still a big enough swing to Labor to carry the seat, but yet again, I observe that pre-polls and postals might change the picture.
6.54pm. There’s been an update that my system has failed to process, so the results are stuck for a short time. Should have new figures in a few minutes.
6.49pm. Now we have two booths from Willoughby where Larissa Penn has thrashed Tim James — along with a third which isn’t nearly as bad for James, but would still be worrying for him in his own right even if it were isolated. But like I say, Zimmerman says scrutineers are seeing something very different on pre-polls.
6.48pm. Some less bad primary vote numbers coming through from the Liberals in Bega, but it still looks like they will be relying on a more favourable dynamic from pre-polls and postals.
6.47pm. Trent Zimmerman on the ABC says scrutineers say pre-polls in Willoughby are encouraging for the Liberals.
6.44pm. Three booths in on the primary vote and one on two-party in Strathfield, a bit of a swing to Labor, but seemingly less than in Bega and Monaro.
6.42pm. We’ve now got two booths in from Bega on two-party, plus six on primary, which means I’m no longer going off preference estimates. The two booths have very different results, but when the primary vote booths are further factored in, it doesn’t look good for the Liberals. But — and here I may end up being a broken record — there may end up being a very different dynamic on postal votes.
6.41pm. So to be absolutely clear about this — my near 100% win probability for the Liberals in Willoughby assumes the Greens will come second, which they won’t. Disregard this figure, now and for the rest of the night.
6.39pm. I’ve managed to gloss over my problem with the Willoughby results page. A second booth, Chatswood Guide Hall, is less spectacular than the first, but still of concern for Tim James. The Liberal-Greens preference throw being conducted by the NSWEC will presumably be irrelevant, the contesting being between James and independent Larissa Penn.
6.36pm. Bermagui and Merimbula Public now in from Bega; Wamboin Hall in from Monaro. Liberal and Nationals still down heavily across the board.
6.35pm. Ben Raue on Twitter points out the number of votes reporting here is about 60% down the equivalent booths from 2019. So on top of all the other cautions, it’s possible the exodus from polling day to postal voting is dominated by conservatives.
6.34pm. Two booths from Monaro — Captains Flat Hall with 189 votes, 183 in Letchworth N’hood Cntr. Very solid swings to Labor in booth. So we’ve got five booths now altogether, all of them bad for the Coalition.
6.33pm. Second booth in Bega — Malua Bay Comm. Cntr — 195 votes, but very big swing to Labor. But I’d certainly ignore what my win probability estimate is saying for the time being.
6.29pm. Very interesting first result from Willoughby — Antony Green is noting this is likely to be her best part of the electorate, but independent Larissa Penn is outpolling the Liberal. I may yet regret the fact that this is the one seat where my results feature might not end up handling the strain.
6.26pm. My two-party preferred will be going off a crude estimate until a few more votes are in. I have a notion the estimate may be flattering to Labor, judging by the fact that I’ve got a bigger projected two-party swing than for the primary vote.
6.24pm. We’ve got a result – 78 votes from Towamba Public in Bega. Meaningless at this stage of course (a solid swing to Labor for what it’s worth), but the good news is that my results facility has dealt with it okay.
6.10pm. Links added above. The results map for Willoughby doesn’t seem to be working, but I’m hopeful it will resolve itself (and this seat is unlikely to be a focus of attention even if it doesn’t), and other than it looks okay so far. However, I won’t have cause for confidence until some actual numbers come through.
6pm. Welcome to the Poll Bludger’s live coverage of the count for the New South Wales state by-elections in Bega, Monaro, Strathfield and Willoughby. If and when I’m satisfied that it’s working properly, I will shortly provide links for my live results pages. First results should presumably be through from country booths in Bega and Monaro around 6.30pm.
/showhide