Friday 5pm. The Liberal lead is now at 878, so unless there’s some late count surprise I’m going to stop following this one.
Thursday evening. As in a number of counts today, the tide has gone out on Labor with 5358 postals boosting Nick Varvaris’s lead from 27 to 789. Still to come are about 4000 absent votes, which behaved similarly to ordinary votes in 2010; 3500 pre-polls, which favoured Labor enough in 2010 to suggest they should recover between 200 and 250 votes; and a little over 500 postals together with 300 provisionals, which are a mixed bag.
Wednesday 4pm. Slow going here, with just 570 absent votes so far added to the ordinary vote count. However, they have had a highly significant effect in breaking 242-136 to the Liberal candidate, which has put him into a 27-vote lead.
Election night. The ABC computer has a 6.8% Liberal swing against a 6.9% margin, and the two-party count of 33349 for Labor’s Steve McMahon to 33251 for Liberal candidate Nickolas Varvaris aligns with this precisely. All primary and pre-poll voting centre results are in.
Poor candidate selection by Labor has thrown away this seat. The pre polls will favour the Liberals hence Barton will have its first elected Liberal MP since 1980.
My seat model wants this one for the Liberals so badly! This one above any other. Please fall. 🙂
Does seem realistic though that having a new candidate would make it harder for Labor to resist a swing back to the Coalition on post-polling. But we’ll see. Labor’s post-poll result last time here wasn’t that bad.
http://kevinbonham.blogspot.com.au/2013/09/2013-federal-election-late-counting.html
Looked at the stuff about whether a seat being vacant helps for post-count swing compared to ordinary votes and found it all disappears in statistical noise and gets confusing. So scrub second para of post above. We’ll see about this one.
As with some of the other seats being examined, this seat has a significantly greater %age of postals cf last time. That is bad news for Labor unless they significantly improve their postal vote performance in this seat relative to the Coalition cf 2010.
Lib now 14 votes in front , will end up about 500 votes margin out of about 12000 votes to count.
For years from now the’refrain will be “who lost the south?”
If non-ordinary votes show the same lean relative to ordinaries as in 2010 then I get a win for Coalition by at least 300 here. But the absents distributed so far showed a more pro-Coalition lean than expected based on 2010, and if that’s a portent then it will get easier.
Absent votes can be highly variable though depending on where they come from, which is often just outside the boundaries of the electorate. This batch may have come from Rockdale or Brighton le Sands, say. So I wouldn’t jettison that 300 vote projection quite yet.
Yes I wonder if I should just ignore absents and assume they will repeat the pattern of the previous election (relative to ordinaries) until proven otherwise. I’ve noticed this effect in a few electorates.
(I didn’t express that very well; what I mean is that I wonder if I should just assume absents will project off the previous election as adjusted for the swing in ordinaries, and not take incoming absents into account until they are mostly in.)
First bunch of postals a shocker for Labor. Looking very hard to hold this.
Second bunch even worse. 643 vote margin now. Still nearly 2000 postals to go.
Gooooooooooooooooorn! Welcome to the new Liberal South of Sydney!
Excellent. (Politics aside.) For anyone wondering about #2, this is the one and only seat tipped to fall by my seat prediction model but not by the betting markets in their insane plunge on the night of September 6.
Kevin Rudd was supposed to rescue the Asian vote for the alp in seats like Barton banks and bennelong but that just did not happen and big swings are happening in those seats. I wonder if the same are happening in melbourne