Close seats – Friday morning update

1. This really has been a very interesting campaign. A combination of self destruction (from the part of the coalition) and a declaration of victory from shorten (no matter what the result is). All combined with The conservative media and elements of the coalition working together to undermine the centrists has really made for good viewing. Australian politics has been more interesting than a hollywood movie!

2. The slow pace of this count is a disgrace to the AEC. There a seats that have still circa 10,000 votes to be counted as at lunchtime Friday…..I have a low expectations when it comes to the AEC…and they are failing even those.

3. @Sandbelter, I take it you’re a proponent of electronic voting?

   In my own case, I am thoroughly torn. On the one hand as a psephologist, I love the time it takes to count votes, it gives me allowance to analyse the results while adding an air of suspense and tension. As a citizen however, the slow count is obviously very frustrating.

   I cannot help but think of how this sorry saga is going to drag on and on with possible recounts and court battles :/

4. I’d rather that we get the right answer slowly than the wrong answer quickly.

5. So, I’ve noticed something interesting about the Hindmarsh count, and it might indicate something about the other seats.

   Hindmarsh got as close as 8 votes, but has since slightly shifted back to 71 votes (as mentioned at the end of the update – except it has moved another three votes since then). This wouldn’t be notable… except that it happened purely because of the later postal votes. No other non-ordinary votes have been counted.

   Considering that the postal votes *were* tilting in favour of the Coalition by about 55% to 45%, this might indicate that we shouldn’t be assuming that postal votes will continue with the current proportions. I don’t know how postal votes get to the counting centres (do the ones that the Liberal Party sent out pass back through them before being forwarded to the counting centres?), but perhaps we’ll see the late postals being weaker for the Coalition (or stronger for Labor).

6. @Glen possibly, although in further counting in the close QLD seats the postal vote has actually trended in the other direction.

   In Capricornia for example, the postals were favouring the LNP by about 54.4%. As of this afternoon, this has grown to about 57.6%.

   Same story in Flynn. This morning it was 62.8% or so and now it is 63.9%.

   Granted in both cases counting is yet to continue and the changes in these seats themselves indicate the volatility when counting. Nevertheless as AG has stated, I would think more likely than not that the Coalition will pull ahead in most of these close seats.

7. I think the 60-odd vote change in Hindmarsh was just a re-check of existing votes, not adding new ones. You see the numbers bounce around by 100 or so quite a bit as the AEC double-checks and finds errors or changes, etc

8. As at 3:00 PM today the races in Capricornia and Flynn have become very tight.

   I’ve also put Cowper in safe Coalition territory – the 2CP lead is now 5920 votes with 53% 2CP counted.

   Capricornia has come in with Labor only leading by 174 with 82% counted. This could still go either way.

   Chisholm is the only Coalition gain from Labor with the gap now beyond 2000 votes.

   Labor leads in Cowan by 534 votes and with only 77% counted.

   At the time of writing, Labor’s lead in Flynn has been cut to just 7 votes. Another one too close to call.

   Forde is also close with the Liberals just 471 in front. Could still go either way.

   Gilmore has the Liberals in front by 1316 and I would think they would be favourites from here.

   Herbert now has Labor in front by only 353 votes with just on 80% counted.

   Hindmarsh is still at 68 in favour of Labor – another too close to call.

   Counting Forde as a Coalition seat, they currently sit on 74. Hindmarsh and Flynn would give them an absolute majority in the house.

   But it’s not over yet.

9. Flynn will likely go the LNP way given previous performances LNP did better in absents votes and prepolls (6-8000 of those to go) +1300 odd postals to go.

   Capricornia will be close but on previous years Labor did strongly on absents.

   Cowan too close to call 534 lead to ALP but LNP has done better in absents in prior years.

   Herbert Im thinking ALP could hold on, just. 353 current lead, line ball when it comes to absents.

   Forde I’d say LNP will keep given line ball absents and current lead of 471.

   Hindmarsh too close to call 50/50, ALP have done better in absents but in those years ALP had a better ordinary.

   What i’ve noticed is the LNP have done markedly better on postals in this election.

   So ALP get capricornia and herbert

   Hindmarsh not sure.

   LNC get forde, flynn, cowan.
   That is from my non computer back of envelope model :D.

10. Could someone tell me how absent and provisional are modeled? Dastayari said on 7 that labor do better but how? Surely the same group of people aren’t absent every election. Its random almost and provisional voters who aren’t on the roll will be different each election. They are likely to be young voters and first timers. Are there any statistical papers out there?

11. @Paul I think you have your answer. Young voters and first timers all favour the ALP – they’re a natural constituency.

    Although I am by no means an expert, perhaps someone has a more ‘learned’ interpretation? XD

12. @Glen, @Mark: I can confirm that the slight drift towards ALP for Hindmarsh has been due to recounting of ordinary votes. There have been no postal votes added to the count since Thursday 12PM (which was when Hindmarsh hit 8 votes)

13. Pollterguy – “Forde I’d say LNP will keep given line ball absents and current lead of 471.”

    I’m confused by that comment, Absents are currently favouring Labor 58-42. Nowhere near “line ball”

14. @ Glen I must of missed that unless they just come in. On that ratio (30% vs 40% with 3500 absents to go) they could make up 450 votes, without taking preferences into account.

15. I think absents is even harder to pick a trend – given unlike postals where they are random, the absents come in from other booths. Its normal to get a decent number from booths just inside another Division (and for Forde if they came from the Centre/ East part you would expect them to favour the ALP)

    That said those initial numbers are definitely positive for the ALP

16. I think the ALP will be very lucky if they get Forde, all things considered. While I see them clawing back ground on absents, I don’t think they’re going to be able to close the gap.

    Cowan and Hindmarsh definitely look very likely for them, with a semi-decent chance in Herbert if the cards fall their way. It is of course very close.

    As the ABC states I think 76 – 69 is the likeliest outcome – a majority of 7, exactly what John Gorton got with similar circumstances in 1969.

17. pick hung parliament still
    labor wins Cowan, Hindmarsh, Capricornia and Herbert

18. Wreathy – 76-69 is not a *majority* of 7. That would be if the Coalition had 82.

19. 76 is a majority of two. 82 is majority of 14.

20. David – a “majority” is when you have more than 50%. While you *could* argue that it’s a count of the difference between the “majority” and “minority”, the natural way to count it is the same way we count the margin in elections – from 50%. If it’s 51 to 49, we don’t call it a 2% margin, we call it a 1% margin.

21. Perhaps the best way of measuring a majority is “how many backbencher malfunctions are they away from minority Government?”.

22. @Glen I am simply stating the method I have seen numerous times.

    Incidentally, it is the one corroborated by Wikipedia. Check out the 1977 Federal Election for example. It is stated that Fraser won a 48 seat majority. Adding up the numbers corroborates this, 86 – 38 = 48 seat majority.

    Furthermore, in every documentary I have seen domestically, a majority is simply how many more than the closest party. The documentary The Liberals for example (it’s on Youtube if you want to verify it for yourself), stated that Harold Holt won a 41 seat majority in 1966. That is, 82 – 41. In 1969, it states that Gorton won a 7 seat majority.

    So if you think your way is better than the way that the ABC used and reinforced by Wikipedia, more power to you 😉

23. @Glen Just to clarify, I am not disputing your method. I am simply expressing surprise since it is not the one I have seen widely used before. I am sure you can clarify this discrepancy for me.

24. Wreathy – the 1977 federal election only had Labor and Coalition seats, so it doesn’t address the issue you raise.

    I think the problem, here, is inconsistent use of terminology. The term “plurality” isn’t often used in Australia, but is a much better term to use for the way you are counting.

    David Walsh’s logic is stronger, as it’s a matter of actual majority – not just difference between the top two. Often, a “majority” will be the difference between votes “for” and votes “against”, with abstentions simply ignored entirely. But you can’t justify using this reasoning with seats, treating crossbenchers like abstentions.

    My argument flows from mathematical reasoning (I’m a mathematician). If there are 11 seats, and you take 6 of them, that’s a majority of 1, because losing 1 seat will mean losing the majority. If you have 5 and the other side has 4, with 2 in a third group, that’s not a majority of 1, because it’s not a majority at all. If you gained one more seat off the third group, you don’t jump to a majority of 2 from no majority at all, you get a majority of 1.

25. To put it another way… if the Greens won Batman off Labor, why would that change the Coalition’s majority?

26. @Glen thanks for clearing that up, I get what you’re saying now.

Comments are closed.