mumble



WWW
 
mumble

home

2007 election links

poll-mix

about

  email: elect AT
mumble.com.au

seat 2pp graphs '07

Seats to watch!

Federal pendulum (old)

margins since 1983

Historical results

Margins 1949  2001

See the lemmings!

two decades of Newspolls

state votes at federal elections

Votes and seat
representation
1949 - 2001

Newspoll
preferences


Newspoll &
Morgan graphs

preferential voting

Newspoll Opposition leader approval ratings

Newspoll Opposition voting intentions

 

 

From Crikey 1 October 2007

7. Which 16 seats should we be watching?

Peter Brent from Mumble Elections writes:

As you know by heart, the ALP is sixteen seats short of a House of Representatives majority, and you can plot those sixteen most marginal government-held seats along the pendulum.

Of course, it doesn't really work that way, and no-one expects the seats to fall in the order they appear on the Mackerras horseshoe. But how might they fall?

I've made my own list of the first seats to fall, or "the seats to watch".

I've started with current margins and performed a few twiddles to reflect my version of reality. For example, everyone would agree that Malcolm Turnbull's on paper 2.5 percent margin overstates his vulnerability because of Peter King's 2004 candidacy. So I gave Malcolm a couple of percent breathing space.

Seats in which members are retiring lose a percent to account for loss of personal vote. However, not Makin, because the retiring member Trish Draper's rorting problems contributed to the 2004 pro-Labor swing, which was the biggest in South Australia. That is, Makin's margin is already deflated.

Seats that changed hands in 2004 get an extra percent to account for the personal vote of the new incumbent, and if it was a sitting member they defeated they get another point because the other party has lost that old personal vote.

Another exception: Labor's Julie Owens gets the one percent for her own personal vote, but not for the loss of defeated member Ross Cameron. Like Draper, his naughtiness produced much of the 2004 swing.

And so on. I factor in a greater than average swing to Labor in Queensland because they all love Kevin Rudd up there, and a lower one in Western Australia because they're rolling in money. (Am I generalising?)

Unlike most observers, I reckon the classic mortgage belt demographic will swing to Labor less than average, rather than more, so I've factored in that, plus a bunch of other vague assumptions, to come up with this list of seats to watch.
[click seat for Antony Green at ABC]

The 16 closest seats on pendulum

My key 17 seats to watch

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

Kingston (SA) 0.1

Bonner (Qld) 0.5

Wakefield (SA) 0.7

Parramatta (NSW) 0.9

Makin (SA) 0.9

Braddon (Tas) 1.1

Hasluck (WA) 1.8

Stirling (WA) 2.0

Wentworth (NSW) 2.5

Bass (Tas) 2.6

Moreton (Qld) 2.8

Solomon (NT) 2.8

Lindsay (NSW) 2.9

Eden-Monaro (NSW) 3.3

Bennelong (NSW) 4.2

Dobell (NSW) 4.8

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

Bonner (Qld) 0.5

Parramatta (NSW) 0.9

Wakefield (SA) 0.7

Cowan (WA) 0.8

Moreton (Qld) 2.8

Makin (SA) 0.9

Braddon (Tas) 1.1

Eden-Monaro (NSW) 3.3

Page (NSW) 5.5

Lindsay (NSW) 2.9

Kingston (SA) 0.1

Solomon (NT) 2.8

Wentworth (NSW) 2.5

Cowper (NSW) 6.7 [but see below]

Stirling (WA) 2.0

Corangamite (Vic) 5.3

Bass (Tas) 2.6

My list has seventeen seats instead of sixteen, because manipulations on the WA seat of Cowan - including the retirement of a popular sitting member - moved it to the government side of the ledger.

The order of my list is very different to the pendulum's, and missing from it are Bennelong (NSW), Dobell (NSW) and Hasluck (WA). That is, I don't think they will be in the first 16 to fall.

In their place are three regional seats Corangamite, in Victoria and the NSW North Coast National seats of Cowper and Page

You might like to try this at home. After the election we can see how close ours lists were.

[Update: the next day, the following appeared in Crikey "letters' section:

"Malcolm Mackerras writes: Re. "Which 16 seats should we be watching?" (yesterday, item 7). I refer to Peter Brent's item in Crikey yesterday. As he notes he has excluded Bennelong, Dobell and Hasluck from his list of the first 16 Coalition seats to fall. Since Peter has invited us to vary that list let me give my alternative. 

As is well known I am predicting Labor wins in all of Bennelong, Dobell and Hasluck. Therefore, I would need to exclude three other seats in Peter's list. The seats I would exclude are Cowper, Stirling and Wentworth. Actually I expect Stirling to go to Labor, but I think it is a less likely Labor win than Hasluck."

Me: Strictly speaking, Malcolm is non-sequiturising a little, as predicting wins in those seats does not necessitate their inclusion in the top 16. (Eg he may be anticipating a big Labor win. [Update: that he is.])

Still, I must admit that the inclusion of Cowper is a stretch. Ok, it shouldn't be there, and I had to try hard to push it into the top 16. It used to be my nomination for the dark horse "ohmigod" seat that goes if Labor wins big time, but these days it is generally recognised as being in play in a landsllide.

If I took Cowper out, I'd put Hasluck in.]

external links
these open new windows

Electoral Commissions
Federal
NSW
    Vic 
Qld WA SA
 
Tas ACT NT
ECA


Parliaments
Federal
NSW
    Vic 
Qld WA SA
 
Tas ACT NT

Electoral

Antony Green (ABC)
Adam Carr
Bullring
Crikey 
Simon Jackman  
Oz politics
Poll bludger
Possums Politics
Upper House

Other

Enrolling the People
Crikey blog
Piping Shrike
Andrew Leigh
Rick Hasen (US)
Oxfam Shop
Australian Constitution
Democratic Audit
Nicholson cartoons
 -
animations

Newspoll

Morgan

WA Uni
election database

Distance Learning