We are now very close to the end of counting for the NSW Legislative Council. The original plan was to finish data entry on Wednesday, with the button to be pushed on Friday morning. This has now been pushed back to Monday morning.
In this post I will run through the votes counted so far (which is most of them) and the possible role of preferences in deciding the final three seats.
This post was originally going to start with an analysis of the votes counted so far and what is left to count. In this regard I built a small model which factored in how many of the votes in the ‘others’ pile had been counted and what was left in that pile, effectively allowing you to match up the two counts and simulate a much higher proportion of the vote counted. That model was in part inspired by the good work being done by Ross Leedham on Twitter. He has been diligently updating his model more regularly to give us a sense of the votes reporting each day.
But so much vote-counting took place across Wednesday that we are now pretty close to the end. Over 90% of the counting has finished in 81 out of 93 seats, with the lowest vote-count being 79% in Newcastle. So at this point I think we might as well just look at the real figures for the check count. These are as of 10pm on Wednesday night.
|Shooters, Fishers & Farmers
|Christian Democratic Party
|Keep Sydney Open
So it’s pretty clear the first eighteen seats have been decided, largely as we expected the day after the election. None of the parties who were lumped in the ‘others’ pile polled well enough to be assured of a seat before the distribution of preferences.
The Coalition has held eight seats, down from eleven, despite not quite making eight quotas on primary votes.
Labor has won six seats, up from their existing five. The Greens have retained two, while One Nation and the Shooters, Fishers and Farmers have each won one.
Then there are three seats in play, with the following parties sitting around roughly half a quota competing for these seats:
- One Nation #2 – 0.5226
- Labor #7 – 0.5090
- Christian Democratic – 0.5029
- Liberal Democrats – 0.4731
- Animal Justice – 0.4198
- Keep Sydney Open – 0.3895
This is where preferences come into play, and things become much more difficult to predict.
Antony Green tweeted this morning that there appears to have been a big increase in how many people are marking preferences on their upper house ballots:
The NSWEC divides votes into three categories: Single Above the Line (SATL) votes which just give a ‘1’ above the line, Random Above the Line (RATL) votes which have multiple preferences above the line, plus Below the Line (BTL) votes. The number of RATL votes has doubled from 15% to 28%, while there has been a 50% increase in BTL votes from 1.8% to 2.7%. This means that over 30% of voters have marked preferences, up from less than 20% in 2015.
This means that candidates will need a vote closer to a quota to win a seat, and gives a better chance for trailing candidates to catch up.
One Nation, Labor and the Christian Democratic Party are clearly the frontrunners for these last three seats. The Liberal Democrats are not far behind, so if they do well on preferences they could catch one of the others (likely the CDP). I don’t have a great deal of information on whether the LDP or CDP is in a stronger position to win.
It seems likely that preferences will flow more strongly amongst progressive voters, whose parties largely cooperated on preferences. This suggests Labor should be in a good position to win.
Animal Justice and Keep Sydney Open are outside chances, but I wouldn’t rule out one of them winning. They both swapped preferences with each other, and received preferences from the Greens and Sustainable Australia. These four parties have about 1.2 quotas left over. Yet many voters for these parties would have cast a vote without the aid of a how-to-vote, and it wouldn’t be surprising if a lot of these preferences instead flow to Labor.
We will find out the results on Monday morning, and I’ll be sure to be back with more analysis after that result.