A full display of the election results, with complete booth figures, swings and probability estimates, can be found here. No updates will appear today (Sunday).
I believe my election results facility, which after a very slow start is now running almost bug-free (with apologies to the seat of Preston), is the only place where you can find two-party preferred results by booth and vote type – vote type being very important in the context of this election. It also trumps the Victorian Electoral Commission site in having swing results at booth-level, on both primary and two-party. To access these, go to the entry page linked to above and follow one of the electorate links further down the page, where you will find neatly displayed results tables with tabs for toggling between totals, percentages and swings. Note that I have turned off booth-matching for my aggregations and predictions, at least for the two-party vote – my failure to do the same with the primary vote means the swings shown for it are slightly anomalous.
The results display turns up ample evidence of what became apparent as the night progressed, which was that pre-polls and postals very often failed to replicate the massive swings to Labor on election day. This meant the final result, as bad as it was, will not be quite as apocalyptic for the Coalition as earlier booth-matched projections made it appear. Cases in point included Brighton, where a 10.2% election day swing that appeared set to deliver Labor a shock victory was followed by swings of only 2.6% on pre-polls and 1.8% on postals; and Hawthorn, where a 10.8% swing had John Pesutto reading his own obituary on ABC Television, only for him to inch to a 53 vote lead after pre-polls and postals only swung 3.4% and 4.5%.
Another seat where the cavalry arrived late was Caulfield, although alert PB commenter Trent notes what is clearly an anomaly in the result. This relates to the booth of Elsternwick North, where Labor has a higher primary than two-party preferred – a mathematical impossibility that can be readily explained by the party’s two-party results having been entered the wrong way around. On this basis, the Liberal margin would appear to be 1.8% rather than 2.9%, although that should be enough for David Southwick after his early fright.
A particularly interesting feature of the result is that the gap between early and election day polling swings was very much a phenomenon of the affluent areas nearer the city. Late counting did little to diminish the swathe Labor cut through the eastern suburbs, which took in Burwood, Mount Waverley, Ringwood and Box Hill. The one exception to this picture was Bayswater, where Heidi Victoria suffered only a 1.5% swing on pre-polls compared with 6.5% on ordinary votes, and ended the night 72 votes in front. In Labor’s other clear gain, the Geelong region seat of South Barwon, pre-polls actually swung quite a bit more heavily than election day votes — 12.6% compared with 7.5%. Nor was pre-poll voting any less harsh on the Liberals in the sandbelt seats, which have delivered Labor stunning margins of 11.9% in Bentleigh, 11.7% in Carrum, 12.3% in Mordialloc and 9.5% in Frankston.
The least unexpected of Labor’s gains was the heavily over quota electorate of Bass, the story of which is told by the suburb of Clyde: in 2014 had one booth which went 376-165 to Liberal, while in 2018 it had two booths that collectively went 1318-932 to Labor. Still in doubt are Bass’s near neighbour, Nepean, where Labor holds a 1.0% lead after swings of 11.3% on the ordinary vote and around 6.0% on pre-polls and postals; Sandringham, where the Liberals now lead by 1.0% and should have the advantage on remaining postals; and Ripon, a Labor target seat where the Coalition performed well above the norm, as they did in Eildon and Euroa, which have in common being regional seats defended by female sophomores.
As usual, it’s been a nerve-wracking election night for the Greens, who are now holding out for their traditionally strong showing on absent votes in a number of seats (to say nothing of the upper house, which I will defer for a later time). This will presumably be enough to keep Ellen Sandell safe in Melbourne, where she leads by 1.2%, and could well allow Tim Read to close his 72 vote gap over Labor’s Cindy O’Connor in Brunswick. However, the Greens have failed to replicate their by-election win in Northcote, and had a rather poor result in Richmond, despite the Liberals making life easier for them by declining to field a candidate. Prahran was a particularly pronounced example of the Liberals doing better on pre-poll and postals voting in inner urban areas, which removed the possibility of their being excluded in what earlier looked a three-way dead heat on the primary vote. So the winner will be whoever finishes second out of Greens incumbent Sam Hibbins (28.3%) and Labor’s Neil Pharaoh (29.6%).
Aside from Shepparton, which was easily retained by Suzanna Sheed, the following seats have independents somewhere or other in the mix:
Mildura. My model says independent Ali Cupper’s 1.3% margin is enough, but it’s not sufficiently cognisant of how well the Nationals tend to do on postals.
Morwell. Labor’s Mark Richards has the edge in the two-party count against the Nationals, but this is redundant as Nationals-turned-independent member Russell Northe will clearly run second. He must then chase down a 34.2% to 20.0% deficit on the primary vote with mostly conservative preferences.
Benambra. The two-party count was Liberal versus Labor, but the potential for interest here lies in the potential for independent Jacqui Hawkins (16.8%) to get ahead of Labor (17.6%) with preferences from independent Jenny O’Connor (12.9%) and then ride home over Liberal incumbent Bill Tilley (40.3%).
Geelong. Independent Darryn Lyons is a clear second on 25.2% to the Liberal’s 20.2%, and Labor incumbent Christine Couzens’ 40.3% is low enough that she might have to worry.
Melton. An exquisitely complicated contest in a normally safe Labor seats which, in which Labor had a retiring member and a late substitute after their original candidate withdrew. There are twelve candidates, most of whom appear to have at least some sort of following locally, and the Labor primary vote is only 34.3%. If preferences from the first eight excluded candidates lock heavily behind independents Bob Turner or Ian Birchall, the might get ahead of the Liberal and then home on preferences. However, one suspects there will be a good deal of leakage along the way. In pure two-party terms, there has been a very unusual 7.2% swing to the Liberals.
Pascoe Vale. Haven’t heard much talk about this, but the primary votes look a bit dangerous for Labor’s Lizzie Blandthorn, who has 37.1% to independent Oscar Yildiz on 25.5%.