email: elect AT

Crikey election book chapter

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
1949 - 2001


Newspoll &
Morgan graphs

preferential voting

Newspoll Opposition leader approval ratings

Newspoll Opposition voting intentions




go to next posts

Nicholson in the Oz

welcome to             



March 23 Mike Rann = results

Someone slipped me Saturday's ALP booth-volunteer's instructions (100kb). (Electorate name blocked out.) Highly prescriptive, particularly the "Mike gets results" bits.

Marky-Mark Mark II

Julia Gillard seems a brilliant person in so many ways, but much of the evidence points to her having a dud political ear. Remind you of anyone? And just as her mentor laid out his 'muscling up' leadership credentials to the Fabian Society a few years ago, Julia did the same yesterday, with (from reports) a collection of straw men and grand narrative cliches right out of the commentator's handbook. She even talked of Labor's need to fight the 'Culture Wars'.

God help us all, history repeats itself.

March 22 SA and Federal: two election results

Mike Rann came to power in South Australia in early 2002 by the skin of his teeth and with a minority of the two party preferred vote. Three months earlier, Federal Labor had lost an election in what was roundly interpreted as a 'flogging' and 'landslide' (until Mark Latham supplied some retrospective perspective in 2004 by increasing the Coalition's majority by 50 percent).






One Nat


Labor 2pp

Coal 2pp

Nov '01









South Aust
Feb '02









As table above shows, the vote results at the two elections - Federal in 2001 and SA in 2002 - were strikingly similar, at least in terms of the major parties, and the two party preferreds almost identical. 

Like Rann, our Mr Howard has survived on minority support, 48.9 in 1998, the lowest winning vote in federal history, smaller than his own losing one 11 years earlier. (See this table.) They're lucky people, today's incumbents.

March 20 Twelve more years!

... of SA Labor in power, so a total of 16. These are the fears of 'senior Liberals'. The same crystal-ball brand as Ms Grattan's (March 14, below.)

Visit any  jurisdiction in Australia today and enquire of the reasons for election results and you'll be told two things: (1) the PM/Premier/Chief Minister is a strategic and political genius with a rare instinct for middle Australia; and (2) the Opposition is a joke.

You and I know coincidences times nine don't happen, and the ever-enlarging wealth of the Chinese middle-class has much more to do with it.

Xenophon: please explain

The big surprise to many was No Pokies MP Nick Xenophon's whopping 21.5 percent vote in the Legislative Council. (The Libs got 25.6). Just as many South Australians had difficulty entertaining the possibility of the biggest Labor majority in history, outsiders didn't get the 'X-Factor'. I'd never clapped eyes on him before seeing him on telly the other day; I can see why the old biddies (always an important Adelaide demographic) would like him: clean-cut, good-looking, a nice boy.

But that's just a blow-in's impression; there's obviously much more to it than that!

March 19 2006 The morning after

It looks like the 29-ers were right about South Australia, or more right than me and my 32. (Labor's seat tally looks like being 29 or 30[?])

However, from reports of the 2pp swing flying around - I've heard 10 percent a couple of times, which would put statewide total at 59 to 41 - it could have been worse for the Libs. 

(South Australia has a rather weird redistribution system. After every election they adjust their electorates to retrospectively ensure that whoever won the two party preferred vote at the recent poll won the most seats. As Labor won government with only 49 percent in 2002, they may have been put a little behind the 8-ball this time. I'm sure Antony Green has written about this somewhere.) [Update: Geoff Anderson has.]

On the Tassie front, Labor's majority in its own right will throw me a couple of hundred consoling Centrebet bucks.

March 18 Super Saturday election day in SA and Tas

SA prediction: Labor to win 32 seats

I'm in Adelaide and will be at tallyroom tonight.

Two opinion polls today, in the Advertiser and Oz. Tiser has Labor's primary down to 37 and Libs on 28, but the paper is repeating some odd behaviour from the 2004 federal poll: not extracting/distributing the 10 percent of undecideds. If we do that for them, we get Labor on about 41 and Libs on about 31 percent. Newspoll has both parties on substantially more than that, but the headline two party preferred is similar: 56 to 44 according to Advertiser, 57 to 43, say Newspoll.

On the prediction front, Charles Richardson in Crikey yesterday went for Labor winning 29 of the 47 seats, as does the pollbludger, while Adelaide's Clem Macintyre says 28.

Adelaide commentators Haydon Manning and Geoff Anderson (who I'll be hanging with tonight) tentatively tip Labor to win 28 to 30 seats. (Well, Haydon does anyway.)

I'm sticking to my earlier 32, which would be a Labor majority of 17 - as far as I know, easily the boldest prediction around.

(However, all the others are more ambitious, as they look at individual seats. I don't do this because (a) I don't have time and (b) whenever I've tried it in the past I haven't done well.)

And for those inclined to believe Mr Rann's warnings that it will go down to the wire, here's the lead of an AAP report on the morning of Victoria's 2002 election, when Labor won 62 out of 88 seats and a 2pp of 58 to 42:

"Victorian Premier Steve Bracks has predicted a close state election today despite polls indicating a comfortable Labor victory ..."

(An equivalent seat proportion today would see Labor win 33. Newspoll's final 2002 survey said  57.5 to 42.5)

In Tasmania, the question is whether Lennon can govern in his own right. That's too difficult to answer, but I have $50 @ 5-1 that says he can, so ... fingers crossed.

March 15 What a difference a poll makes

Yesterday's Newspoll slotted oh so perfectly into the fortnight's narrative, multiplying Beazley's woes and consolidating the likelihood that he'll be gone by Xmas.

I often complain about Newspoll's preference allocation, and I do again today; this time it's Labor they're too kind to. The Coalition on 45, Labor on 35 and Greens on 6 is more likely to round to 54 to 46 after preferences than 53 to 47. The difference is deeply trivial in every way - except that 46 percent is the 1996 wipeout  rather than the 2004 flogging, a much more dire hook for yesterday's wall to wall reporting. The frenzy from just one opinion poll was absurd, but it could have been much worse.

We've seen blood in the water before, but the difference between the Crean's white-ants and Beazley's seem to have been that one was based on internal polling allegedly showing Labor heading for a drubbing, while the other was because while the polls were ok the leader 'wasn't sorting out the factions', 'wasn't inclusive enough' or 'wasn't trying hard enough'.

Anyway, the odds of a Beazley knifing before the next election are surely now better than even; let's hope they make the sensible replacement choice this time.

March 14 Newspoll in the Oz

[Irony on] A towering figure who attracts great reservoirs of respect and affection within his party. The rank and file recall with gratitude how, during his all too short leadership, the plucky fellow took on the factions and won. Ordinary Australian men and women saw in Simon a leader with gumption - not dashing or charismatic, nor the avuncular Beazley, but something much, much more: a conviction politician, a man with ticker. He is deeply popular, and Kim forgets this at his peril. [Irony off]

Back on planet earth, clever Lemmings have returned Labor to Boofhead's election result numbers.

Victoria: something in the water?

Michelle Grattan looked to the future in The Age yesterday:

"Few, least of all in the Labor Party, doubt the Coalition is likely to win again next year (even if Costello leads it). Some are starting to worry that it might be able to increase its majority. And that, for Labor, would be disastrous, effectively guaranteeing it would lose the 2010 election too."


Her colleague, Shaun Carney, rarely fails to advise Labor supporters to slash their writs over the prospects of ever seeing a Federal Labor government again, and then there's Andrew Bolt over at the Hun. Imagine reading these guys every day and you can see why Simon, Julia et al have gone bonkers. But did columnists drive them mad, or they all a symptom of something else? 

What is going on down in Melbourne?

March 9 More ACNielsen 2004 numbers

So number-heavy, in fact, that I've moved it off page. Click here for post.

March 8 2006 Dear Kim: 'Neither a Lemming nor a Rooster be'

Mr Beazley seems to have forgotten this golden rule of Labor politics, learnt by generations of young 'uns on the knees of their Elders.  Lemmings are human too and need to feel loved. Would we have seen this week's Lemming eruption if Kim wasn't so Rooster-friendly? 

On the other hand, I was hoping for Crean's survival, but his behaviour since has been almost delusional.

Can't they just de-select all the spoilt brats above and put Rudd in as leader? (Or better still Peter Beattie!) Would make for a peaceful life. What they say about disunity is true: remember  Andrew Peacock having a certain shadow minister named John Howard yapping at his heels during the 1990 election.

South Park nonsense Pt II

On Caroline Overington's account of kiddies voting for Howard last week, Tim Colebatch from The Age sent in these Nielsen numbers, taken over the 2004 campaign, total sample 'more than 6,000' and the 18-24 sample size 665 (so error margin a little under 4 percent). The aggregate two party preferred came to the exact election result(!); these numbers have youngies, as you would expect, overwhelmingly voting Labor, and actually swinging to Latham (by 1.1 percent) in 2004.

            Lib/Nat    Labor   swing from 2001
18-24        45.4       54.6          -1.1
25-39        49.0       51.0           0.5
40-54        52.6       47.4           2.8
55 +         59.3       40.7           0.2
total         52.7       47.3           0.9

(The aggregate swing of 0.9 isn't big enough; will have to find out about that.) [Update: swing from Nielsen's 2001 numbers, which overstated Coalition support by about a percent. More from Tim]

Tim reckons the AES is 'consistently overweight on upper/middle class respondents'; as noted here before, the 127 sample size was way too small and they always(?) over-state the winner's margin. Unlike Nielsen, Newspoll, Morgan etc, AES isn't weighted either (as I understand it.)

(More above, March 9)

Go to posting on Overington piece.

Continue to previous posts


external links
these open new windows

Electoral Commissions



Antony Green (ABC)
Adam Carr
Simon Jackman  
Oz politics
Poll bludger
Possums Pollytics
Peter Tucker
Upper House


Inside Story
Enrolling the People
Crikey blog
Piping Shrike
Andrew Leigh
Rick Hasen (US)
Oxfam Shop
Australian Constitution
Australian Policy Online
Democratic Audit
Nicholson cartoons
Online opinion

WA Uni
election database

Distance Learning

Yarrh wines