October 15 ACT election notes: the view from Canberra

Some uninformed jottings ...

Labor has more telly ads, as incumbent governments tend to. They don't mention Chief Minister Jon Stanhope, and instead go on about the wonderful things 'ACT Labor' has done and will do. Also negative ones about the Liberals: they are a rabble, can't govern themselves etc.

Liberal ads barely mention their leader (Zed Seselja) either, but instead resemble those for a bunch of independents: 'I'm Joe Bloggs, I'm a great bloke who'll do great things for the electorate (oh, and I'm a Liberal)'. There's one with a yapping Stanhope's head, getting kicked. Labor, a la federally in 2007, has a response, with that ad on a telly and deputy Chief Minister Katy Gallagher (presumably the government's most popular member) in the foreground tut-tutting about the opposition getting 'personal'.

What's gonna happen?

To be honest, I haven't gotten my head around the likely permutations in the three electorates. Obviously Labor will lose its majority. 'The suburbs', meaning Brindabella and Ginninderra, seem to be where the action is.

But I expect more than that: I reckon Labor will do 'worse than expected'. Not because I live here and have a feel for things, but mainly because that's what happened in NT and WA. The knee-jerk (and probably incorrect) explanation for those unexpected results had to do with governments 'going early'. That one won't be available for this result.

It might be my imagination, but Labor ads seem to contain an element of panic. (Or they're just not taking any chances.)

Still, Canberra is a Labor town. Stanhope is reassuring people that his government will look after people hit by the economic downturn, which surely resonates. 

It's hard to see how a Liberal government could be formed, where the numbers would come from. This is what the last election that resulted in a Liberal government looked like, supported by the Osborne Independents (Paul Osborne was a former footy player.)

But Centrebet is paying $4.15 for a Liberal win. I reckon it's more likely than that, so have thrown some money on it.

• See also Antony Green and Pollbludger here, here, here and here.

October 14 Newspoll says 55 to 45 

In the Oz, here. Note the low primary votes for both sides (41 v 38), and high Green one (of 13). Turnbull's approval a high 50 percent.

October 13 Commentator in cuckoo-land Caught Andrew Bolt on 'Insiders' yesterday predicting that if John McCain loses this presidential election, Sarah Palin's trajectory will continue ever upward and she will be the most likely Republican candidate in 2012. (Transcript not up yet.)

What an utterly preposterous prediction. Palin's only future in national politics lies in a McCain win (which I don't consider as unlikely as most). 

If they lose, Palin will get a lot of the blame (although not as much as she deserves - or, rather, the dills who chose her). The scales will fall, and there will be no shortage of conservatives declaring they always thought it was a dreadful mistake.

She will return to obscurity - in the national political sphere at least. Perhaps a talk show beckons.

The echo-chamber

What makes commentators say such la-la land things? Much of it has to do with the kind of blog/journalism that Bolt practices, where everything is web-based, the ambit covers any topic under the sun, "research" is via google and links from like-minded sites, and - most importantly - adoring fans agree with your every utterance. 

If you're stuck in that echo-chamber, you can start believing anything.

October 11 Two elections and their lessons

Two countries that are 'similar to ours' are having elections soon. 

1. New Zealand: no lessons for Australia

New Zealand's is next month, and they are going to do what Australia did a year earlier: replace an old government (nine years in their case, 12 in ours) with a new one. 

It's fair to expect that, as here, the result will be somewhat closer than the opinion polls (which have the Nationals over 50%) have indicated. Indeed, Morgan's latest has a narrowing. But that the Nats will win is not in doubt - although who they find to form coalition with, if they don't win in their own right, is an interesting question. 

NZ will have no lessons for us; in fact we showed them what to do 11 months ago.

2. Canada: a hint to the future

By contrast, Canada's one-term government going for re-election amidst world economic turmoil might provide some clue as to how Rudd & co might fare at the next election.

Centrebet did have a win for the opposition Liberals paying $10, but that has narrowed to $5.25 (against $1.14 for the Conservatives). Presumably that means opinion polls have narrowed somewhat, but expectations are still obviously lopsided

The Conservatives had a narrow win last time, but that's all but irrelevant now. The fact of incumbency and its length are the important factors.

If Canada's Conservative don't win comfortably on Tuesday, that should be worrisome for Mr Rudd (and encouraging for whoever is Liberal leader at our next election), because it could indicate we're moving from an era of long term governments to an era of short term ones..

Of course, the future worry for Rudd is a recession, and Canada, like Australia, is not in one yet. Maybe if Harper wins big time, the lesson for Rudd is: go early, before the economy tanks.

October 9 Centrebet's McCain payout blows out to $4.80

This seems an extreme interpretation of what is only a five point or so lead in national polls. 

For context, recall that Rudd was regularly at least ten - and often more than 15 - points ahead of Howard throughout 2007, right up until a few days before the election. But the odds never blew out like this.

My bets

As you know, I had been pleased with myself for getting $3.50 on McCain. But am now a little relieved at hedging much of it in a late September punt on Obama @ $1.60.

As you also know, my feelings about this election changed with McCain's choice of running mate - but in the opposite direction to most.

Am now sitting on fence ... for time being. Either side could win.

Track evolution of thoughts on Palin here (day after the announcement), then here, here, here and (most importantly) here.

October 4 Palin v Biden debate Sarah Palin is obviously a very accomplished performer; she delivered lines mostly with ease, mangled words only a couple of times, even somehow got away with a couple of winks to the camera. Possesses oodles of charm.

But a 'narrative bullshit' moment: her 'there you go again, Joe', an obvious reference to 1980 Reagan v Carter. What for? Who for? Just reminds everyone how scripted these events are.

Palin easily cleared a very low bar and Republican supporters everywhere must rejoice. 

But she is still, from McCain's point of view, a terrible choice for VP.

Senator Biden's smile If a person viewed this debate without preconceptions, they would probably judge Palin to have won the charm contest, while Biden obviously knew lots more about everything (although he managed to generate some charm too).

But debates are not just about who knows the most, but the whole package.

Biden often smiled too broadly. (Are those teeth capped?) One suspects he has a 'smirk problem', and his orders are that whenever he finds himself grinning at all, he should go the whole hog. Results were often inappropriate.

Same syndrome has seen our own Peter Costello pull some silly faces over the years.

October 3 Wanted: one senior public servant responsive to minister's needs  statutory independent officer

Several weeks ago, the Special Minister of State, Senator Faulkner, announced ever so quietly that the Australian Electoral Commissioner was leaving three years into his five year appointment, to head the department where he had previously been deputy secretary.

Last Saturday's Weekend Oz advertised the position, and if you think running the AEC is something you'd like to try, here are selection criteria etc. Apparently it's the first time an outside recruitment agency has been used for this position.

I think Ian Campbell was the first commissioner with no electoral background; electoral experience is also absent from this job description.

October 1 Sarah Palin & Mark Latham - the similarities

This is an elaboration of a comment made several weeks ago.  What, you might ask, can these two possibly have in common?  I can think of a couple of things ...

The logics behind the installations of both were similar. Each was going to be a circuit-breaker, give 'definition' to a bland brand, 'cut through' to the nightly news. Both in the 'just so crazy it might work' genre. 

'Conviction politics!'

They were both about attracting votes, without consideration of the ways they would repel them. It is generally forgotten that Mark Latham was (before the 2004 election) a very popular guy. He attracted the then highest approval ratings for any opposition leader. (Was later overtaken by Kevin Rudd.)

Latham's approval started going south before election day, but even his last Newspoll pre-election satisfaction rating election of  54 percent was higher than John Howard ever attracted as opposition leader.

But Latham, although much liked and admired, was hard to vote for, and delivered the worst opposition performance since 1977.

John McCain's drawcard used to be believability and seriousness, no gimmicks, as against the flaky celebrity on the other side. No more.

Yes, Americans are different, but they're not that different. As a presidential candidate, Sarah Palin would be unelectable. And as VP, she might just be.

Revenge of the speechwriters?

There's another way of looking at this. 

In all aspects of life, there's reality, and then there's the way it is represented, in digestible form, in three acts. We all need the latter to make some sense of the former.

Speechwriters exist in the second world. They are there to give things meaning, to tell stories. 

But you wouldn't plan a battle strategy based on the film 'Gallipoli'. And you wouldn't base tactics on the sort of stuff that even the greatest speeches contain.

'Fight them on the beaches'? Best check with the generals first.

So perhaps appointments such as Palin and Latham are what happens when speechwriters are given tactical responsibility. When people believe the stories. To slightly misquote (former speechwriter) Peggy Noonan, they represent the 'narrative bullshit'.

There is, sadly, a reason the species politician has evolved into a grey and boring equivocator, trying to be all things to all people when running for office. That's what works.

Sept 28 'Triangulate', young man 'Triangulation' is a horrible word that usually comes out of the mouths of spivvy political consultants of the left.  I think it means centre-left candidates should from time to time say 'right wing' things, to mix it up, make the message more complex and durable and avoid looking like a captured, tick-the-boxes lefty.

Which is true, but it hardly warrants such a pretentious carry-on. 

(Our own Mark Latham was rather prone to gimmicky jargon such as this.)

Having said all that ...

Barack Obama has already apparently enraged some true believers by moving to the centre on some issues. After watching the presidential debate yesterday, I reckon he could also, for want of a better expression, 'triangulate' more to give resonance to his 'post-partisan' shtick.

(John McCain, with his positions on torture, past 'agents of intolerance' utterances etc already embodies the concept.)

And rather than boasting 'I was right on Iraq five years ago', how about: 'yes, I was right on Iraq, but that's not important right now, it's what we do now ... '?

I'm probably not saying much that hasn't been said already.

Friday's vice presidential debate should be a cracker.

Sept 26 Mr Morgan says 57 to 43

Here. He draws attention to Rudd's dissatisfaction rating being higher than Turnbull's, but that's largely a function of the high "don't know" response - par for the course for new leaders - of 33. Given this, Turnbull's 43 percent satisfaction is quite impressive (if you believe in such ratings).

The undecided 33 will presumably about halve in coming months; which side they go to is the question.

(This week's Nielsen also had a high approval for the opposition leader (given the high undecideds), while Newspoll, oddly, gives Rudd's approval but not Turnbull's.)

United States election: A 'close' result?

The general expectation seems to be that November's presidential election will be close - it will come down to a few States. Greg Sheridan expressed the consensus view recently:

"It will be a very close election. The overwhelming majority of states will vote the same way (that is, Republican or Democrat) as they voted in 2004 and 2000."

It depends what 'overwhelming majority' means, of course. If one candidate wins all his party's 2004 States, plus five more, he will comfortably become president, possibly in a (depending on which States they are) 'landslide', even though 90 percent of States voted the same way as last time.

And if we allow our 'only a few States in it' to be the biggest three - California, Texas and New York - that covers a huge range of outcomes. Eg if Obama wins, say, 350 of the 538 electoral college votes, you could still say: 'McCain would have won if he had just taken two more States - New York (31) and California (55)!' [I'm willing to be corrected on the numbers.]

Still. Let's assume what folks like Greg mean is that only a few States will change allegiances from 2004, and that the result will come down to a reasonably 'in-play' State or two.

Without great macro knowledge of this election, this strikes me as terribly unlikely. And so does the prospect of any kind of 'close' result. The fact that 2000 and 2004 were two of the closest in history doesn't make a close one this time more likely. 

People also reckon it'll be close because that's what the State by State opinion polls say. But with the national ones jumping all over the place in recent weeks, surely the State ones are too. And many must be too close to call.

I reckon it's partly wishful thinking: we'd all love a great nail-biter on election day/night (ending of course with our favoured result). It's also an equivalent of our own 'hung parliament' syndrome. 

And it's an unrealistically static view of voting behaviour. Surely if there's one thing we can expect of this election, it's lots of people who voted for Bush last time going for Obama this time, and lots of Kerry voters supporting McCain. Some who voted last time will stay home in 2008, and vice versa. These will spill into State totals.

And if one side wins the national vote by, say, 5 percent, which could easily happen, it will probably translate into a huge win in the electoral college. And if the national vote is much closer than that, that too is most likely to translate into a comfortable win.

Uncertainty about the result should not lead us to expect a 'close' one.

[As you know, I was expecting a McCain win until the blunder. Now not sure.]

Sept 25 Queensland and South Australia

Here are the Newspoll tables for Qld and SA. It's fair to say there will be big swings in both those States next year/2010, with the only question being: will they be enough to change government? My guesses are: 'yes' and 'perhaps', respectively.

Sadly for the federal government, NSW voters cannot vent their elephantine collective spleen until 2011, by which time they will be very cross indeed. Luckily for those NSW folk, however, they'll be able to unload on the next best thing - Kevin Rudd - before then.

And Victorians don't go to the polls until late 2010.

Sept 24 Newspoll shows Labor slumps in SA and Qld

In the Oz. Down to 50 50 and 51 49 respectively, assuming Newspoll is allowing for the 'O' in Queensland's OPV, otherwise we can call that 50 50 as well. (See recent NSW Newspoll.)

State Labor's on the way out; John Howard was the best friend they had (and they were his).

(References here and here to Labor 'needing' Green 'preferences'/'support' to form government in Qld are trivial. If federal Labor hadn't received any Green preferences last November, John Howard would still be Prime Minister. Ok, that statement, while true in some strict senses, was a bit of a sleight of hand, but that's discussion of preferences for you.

See last year's results under first past the post and optional preferential.)

Sept 23 Candidates' lists?

Has anyone seen lists of names of who voted for whom, either/both for last November and last week? Would be interesting to see, for example, who changed their votes which way.

(Like the 'Lemmings for Beazley' exercise.)

And another thing ...

Note that Nielsen's 2pp estimate based on preference flows at the last election makes it even closer than the headline numbers: 51 to 49. From this.

Sept 22 Turnbull v Rudd: Two different polls

Nielsen says 52 to 48 (from 41 to 42). Newspoll 55 to 45 (from 42 to 38).

[Update: Mr Morgan will also be in dipping his instruments as well.]

The double-breasted youngsters in the PM's office will be all a-panic this morning. "Ok - this week we really hit them with policy announcements!"

A post-it note suggestion

Bill Clinton, as 1992 presidential candidate, famously had "It's the economy, stupid" stuck on the wall to keep the troops focussed on the job at hand.

Mr Rudd's office could, perhaps, do with something like "We're in government now."

Update: Opinion poll tables

Here's the Nielsen.

Here's that weekend Galaxy and one from March for comparison.

Sept 21 'Insiders' - out of touch elites?

A visitor to this country who, fresh off the plane on a Sunday morning, switched on the ABC telly at 9am, would get the impression that the Labor government, only just elected, is in diabolical trouble. The Prime Minister in particular is always making a hash of everything.

They must be doing terribly in the opinion polls.

This morning, for example, the 'Insiders' crew all agreed that Kevin Rudd's announcement of a local government summit in a couple of months is a terrible idea. Why? - not sure, something about it not striking the right tone, or something.

The three tiers of government in Australia are out of whack. In other countries, local government has a greater everyday presence in people's lives. It is (I think) the one tier we could abolish without a referendum. This is an important issue.

But, no, our commentators reckon that the Wall Street melt-down should have elicited an economic announcement. Mr Rudd should have inserted himself in the 24 hour news cycle with a gimmick of some sort.

Who is it who is supposed to be all spin and no substance again?

Sept 20 Opposition leaders' time in office - two tables

Further to the other day's Oz piece about Malcolm Turnbull. Table below ranks federal opposition leaders this side of December 1972 by ascending time in position. Months are approximate only.

1. Federal opposition leaders since 1972 - months in position

* Won the election

Apart from Downer, who didn't contest an election, and Peacock II, all the top spots are occupied by the election-winners.

The second table is a little different, showing months from their installation until the election. So Downer, Crean and Beazley II aren't there, and Peacock I, Howard I, Hewson and Latham's post-election months are excluded. Bomber I contested two elections, so appears twice.

2. Federal opposition - months in before election

* Won the election

In this as well, the election-winners are all in the first column.

I place Boofhead with Rudd, Howard II, Peacock II and Hawke as leaders installed five minutes before highly winnable polls. They should have had the world at their feet, as an election on the immediate horizon guarantees unity - until election day anyway - and a bounce in popularity.

Peacock didn't quite get the unity, thanks largely to a famous 4 Corners episode. But Latham certainly did (until election day).

Assuming the next election is three years after the last one, Turnbull has to match Sneddon in overall longevity. And be the longest serving opposition leader since Whitlam (1967 to 1972) to become PM.

Things were different in the '60s and '70s. Today the obsession and frequency of opinion polls makes life very difficult for federal opposition leaders.

That's why just surviving until the next election (unless it's early) is a huge ask for Malcolm. Let alone winning it.

Sept 19 That 'n this

1. Penny drops that the Pollbludger is using, as the mug shot for his new spot at Crikey, (a cut-out from) a snap I took of him at the National Tallyroom in 2004.

My lawyers have started the clock.

2. Oz Politics is back in action.

Sept 18 Grand old man of politics It is with some sadness that I have taken Dr Nelson down from the top spot, above. But am yet to find an appropriate replacement picture. So here is the stop-gap: the Father of the House*, commanding deep reservoirs of respect from all sides of politics.  When Wilson rises, the House falls silent.

Nicholson in the Oz

*As several pedants have pointed out, that's really Philip Ruddock. I'm bending the term a little.

Sept 17 Dear Malcolm: 26 months is just too long

I'm in The Australian today, largely elaborating on yesterday's post.

Here.

US Presidential odds

Someone's been putting money on Obama again, because his Centrebet payout has, after doing nothing but rise since the Palin nomination, now slipped from $1.70 to $1.65. [12:30pm update: $1.62]

Probably due to the tumbles on Wall Street.

Sept 16 Update: Peter Costello's day? Malcolm Turnbull's chances of ever becoming PM just took a dive, and Joe Hockey's and Julie Bishop's just improved a little.

If Peter Costello is still preening around parliament beyond the end of this year, we will know he covets a last-minute change before the next election.

And if the world slips into recession and the government's popularity dives, it might just work for him. (The last one-term federal government was flattened by the Great Depression.)

In any event, there is little chance of Malcolm still being leader at the next election (unless it's early). 

Federal Liberals: 'New leadership'

I think that's what Dr Nelson is promising. Apparently he is favoured to hold on today, which only makes sense in that if he's willing to keep taking the hits for a while longer, it is rational for the party to let him.

They can cut down his swinging carcass any time. He certainly won't be contesting the next election (unless the government springs an early one).

Clearing the air?

'Circuit-breakers' don't really happen, and if he wins today, Brendan will continue to live and die by opinion polls.

Newspoll will still arrive every fortnight. Drip drip drip.

My bets

Below are Sportingbet's odds for today. I've thrown a tiny amount on Turnbull, and the same on Bishop. I've probably done all my dough; my Bishop reasoning is here.

Sept 15 WA horse crosses the line By a nose.  Phew.

ACT the next race

The next Australian election is in the ACT on 18 October. The electoral system there makes it tricky, but Canberra's pro-Labor tendencies mean it is reasonable to tentatively anticipate, as most do, a Labor win. But certainly not another majority one.

However, in case it means anything (which I doubt), ACT swung to Labor by the smallest amount of any state/territory at last year's federal election - ie less even than WA - as this table shows.

State/Territory two party preferred federal ALP votes 2004/2007

Electoral finance green paper

Where is that green paper? It was due in July. Stuck somewhere in Mr Rudd's office?

Sept 14 The boys finally move

After some carry-on, Pollbludger and Possum have shifted to Crikey - here and here - where they have lots more techy stuff at their disposal. Should be interesting.

The next federal Liberal leader (but one)?

If the Sportingbet odds below were for who will be opposition leader at the end of this year, they would be enticing. The Turnbull and Bishop amounts roughly reflect their chances: it'll probably be Malcolm, but there's an outside chance of Julie.

But for leader at the next election, none of these options looks worth it. (And what is Dr Nelson doing at $2.50? He'd be unbackable at $100.) I operate from the belief that whoever replaces Brendan next week/month will probably not be around for the next election (assuming the government runs a full term). 

Even the brilliant, self-actualised Turnbull would finally meet a situation beyond his control. Larger forces are at work here than mere individuals.

The possible permutations before the next poll are therefore too many. It's too unpredictable. Maybe that makes Joe Hockey and Tony Abbott the best value bets. (Well, Joe anyway.)

Sept 13 US president bets

This is what they look like at Centrebet this morning. They're edging ever closer; just last night Obama paid only $1.62.

I could hedge now - for example, having bet $X on McCain @$3.50, I could put 2x$X on Obama @ $1.65, so having bet a total of 3x$X, would be ahead whoever won. 

But we can expect the amounts to continue narrowing for a while, so will wait.  [Update Sept 14 2:30pm, $1.70 to $2.10.]

(One risk is the absence of an 'any other candidate' option. There was one earlier with long odds, which I took out as insurance. It's not beyond the bounds of possibilities that one of them will not be there in November. [Update: but Portlandbet has the 'other' option.])

On a certain Liberal's leadership ambitions Mr Costello discussed the Hawke/Keating 1991 leadership change analogy with Mr Hartcher this week. Said Peter (quoted by Peter):

"Keating was very, very lucky in this respect - he had John Hewson." Against any other Opposition leader, argued Costello, Keating would have lost the 1993 election.

"The idea that you can destroy a government then win an election depends critically on a John Hewson walking down your path. There was no John Hewson about to walk into the Labor Party."

Calling the post-Hawke government 'destroyed' is a bit of a stretch; traumatised, perhaps. Costello is no doubt giving a whack to both Keating and Hewson, who have said unkind things about him. 

But of course a John Hewson did 'walk into the Labor party', and his name was Mark Latham, for the 2004 election. 

But that's also where the Hawke-Keating comparison falls down. When Hawke was toppled in late 1991, his government was obviously in trouble. Toppling a leader who seems to be doing well is something else, and Howard was only obviously (to the average observer) vulnerable in 2007. 

Costello would have defeated Latham in 2004, no doubt. But I suspect what he's really saying is that toppling Howard in 2007 (if it were possible) and then going on to lose the election anyway, would have ruined him. He would be widely despised in his party.

Can you blame him? As things turned out, he looks like bowing out a very happy chappy.

('Looks like', because we still don't know when he's leaving. Let's not fully rule out this scenario: this time next year, the economy is in terrible shape, the government is unpopular and Peter is still in parliament. He then topples whoever is leader and contests the next election.

Or this variation: whenever the next election is called, including if it's a DD, Costello topples the leader. Then he does what we can now call a 'Hawkie/Barnett'. )

Sept 11 2:05pm Movement at the station - but where to?

Pollbludger and Possum are up to something. It seems that this and that link will, some time this afternoon, cease to exist and instead forward to something new and exciting.

?

[Sept12 Update: crikey(?) they're dragging it out.]

[Sept13 update: it's finally happened; they've both moved to Crikey. Here and here. ]

US Presidential election: Republican poll boost

Level-headed observers (emotionally committed partisans need not apply) are waiting to see how the Palin nomination pans out. Previously I thought Obama had no chance of winning, but that changed. In one swoop, McCain trashed his most valuable credentials.

But the Democrats seem to be into the 'narrative bullshit' too. Or it might just be a matter of time before they flick to scary.

Just glad I got my $3.50 wager. At last count it's $2.35 and dropping.  

I'll be hedging when the price is right. 

Sept 9 On the National Party as kingmakers

To make an obvious point: if the Liberals and Nationals did not always form a Coalition when between them they had enough seats to form government, the federal elections in 1998, 2001 and 2004 would have thrown up 'hung parliaments', making the Nats 'kingmakers'. (The same for most conservative victories over the last century.)

They would have ummed and aahed between Liberal and Labor, and then made a choice.

People like Ross Fitzgerald advocate more independence for the Nats, but is there any future in that? If they do support a WA Labor government, it would most likely be very unpopular, splutter along for a while before falling over, at which early election the Liberals would easily win.

And how might the Nats do in that poll?

On the other hand, if Grylls is seen to squeeze a good deal out of the Liberals in return for support, what position does he adopt at the next election? Does he pretend he might well - you never know - support a Labor government?

And, finally, did running as a third party on Saturday maximise or minimise the total number of conservative seats? We don't know the answer to that; if the latter, and Carpenter does fall over the line with the two independents, the Nats' self-indulgence will have cost the conservatives government anyway.

Perhaps this show of 'independence' is a gimmick you can only pull once in a while.

Disclosure: the writer has dosh on the Libs forming government.

Sept 8 pm: WA two party preferred about 48.5 to 51.5?

If I understand Antony correctly, we won't get 2pps at ABC, although eventually, I imagine, the WAEC will publish some.

From these ABC numbers at left, I reckon the state-wide two party preferred is about 51.5 to 48.5 in Libs/anti-Labor's favour.

That is: From this, the Christian Democrats, Citizens Electoral Council and Family First, right-wingers all, account for about half the 'others'. At an approximate guess we could give Labor about 9.5-10 percent of the Green vote in preferences, and about 3-3.5 of the 'others'.  Maybe closer to 3.

Which comes to about 48.5 percent. Or a little more. Very roughly.

(Of course, I'm treating the Libs and Nats as a coalition, which one must for these purposes.)

Newspoll: credit where due

Check out Newspoll's primary numbers. Very close indeed.

(Don't have Galaxy data but will put out a call for them.)

[Update: Here is Galaxy's. Favours both major parties overly, but the two party preferred looking spot-on.]

September 8 The post-WA future

Here come the Reasons. (Actually, they began began before Saturday, when public and leaked party polling showed a swing on.)

It's the fault of the ALP campaign, they say: too negative, anti-uranium message was a dud, all over the shop etc. Arrogance, cynicism, Premier Carpenter put in a shocker.

Like the Libs federally, over a period of three years WA Labor somehow forgot how to campaign! State Liberal HQ put in a blinder.

Let's rise above all that post-hocery and check out the forest.

We know NSW Labor will lose in 2011, and the Northern Territory at their next poll. I now reckon Queensland and Tasmanian Labor are probably gone too. Maybe Victoria as well. South Australia - last on, last off - is the only one I strongly favour to hold on at their next election. But watch out for a big swing there too.

Which all sounds glib and absurdly deterministic, but ... wait and see.

History repeats

When John Howard came to power in 1996 there were conservative governments in power in every state/territory except NSW. By 1999 three more Labor governments had been elected; by 2002 it was wall-to-wall.

Here we go again.

[Malcolm Mackerras wrote in the Canberra Times last month:

"During the eleven-and-a-half years when John Howard was prime minister there was, literally, no case of a state or territory general election at which the Liberals performed better than expected ... "

But, wrote Malcolm, in recent years he would tell Libs to look on the bright side because: 

"as soon as Howard was gone, the reverse pattern would emerge. In the Northern Territory on August 9 we had the first test of my theory that the Liberals would now perform better than expected. The Western Australian elections on September 6 and the ACT elections on October 18 will be the second and third tests."

I'm on the same page as Malcolm.]

September 6 WA: Newspoll in the Oz says 50 50

From 35 to 43, 12 and 10 other. See tables.

Word is that Newspoll has seen the light and is now estimating reasonable preference flows. That appears the case here (eg give Labor 10 of the Greens' 12 and half of the 'others'); looked at another way, either side could come out ahead from those primary numbers, so 50-50 is a good summation.

My hunch is that the Lib/Nats will win the two party preferred vote today. But will that be enough?

My weaker hunch is: yes.

My weak tip: a Barnett premiership.

So I'm tipping, with no great certainty, a change of government in Western Australia.

(We must assume the Nats will support Barnett.)

More ripping yarns from the US of A

Heard an American commentator on radio yesterday suggesting that, given that the circumstances at this election should lead to a Democrat win, it seems the Republicans have nominated the one candidate who could possibly win, and the Democrats the one person who can't. 

I agree with the first part, but not the second. As you know, I think Obama will have great difficulty getting elected, but the same would have applied to the other main contender, Hillary.

About Palin

Here's the thing about the Sarah Palin veepness. It does indeed exist in the realm of 'narrative bullshit' - of punching through, making a connection, causing a seismic shift, creating momentum. 

It is certainly exciting commentators who can't see past today's headline, but in a month's time both conventions will be forgotten and we'll be back, in the main, to the two leaders.  

I think everyone would agree Palin would be utterly unelectable as a 2008 presidential candidate. But as a VP one, part of the show that's gonna give Washington a (modest) shakeup, she has, under the steady hand of John McCain, knockabout appeal, youth, contrast etc. (Assuming nothing too nasty emerges from Alaska.)

So the challenge for the Dems is to make her the presidential candidate. If they can do that, it becomes a contest between two negatives. Flow-on benefit = questions about McCain's judgement.

September 5 Two photos You know what I'm saying.  Does anyone know the statistical chances of a 72 year old making 76?

[Update: I'm obviously behind the eight-ball on this, as several readers have sent the answer, which seems to be about 82 to 85 percent. Links here and here.

Tricky thing for the Democrats to play, of course.]

September 4 WA election two days away

If I do say so, I had a great WA election in 2005. I was one of the few people on the planet to consistently predict a Labor win, and along the way I made new friends and took some Centrebet dosh.

Three and half years ago, Paul Murray predicted a Coalition win, but as Pollbludger (who you may have caught on 7:30 Report last night) says, this time Paul is going for a  Labor win. (Or he's trying to combat their tedious 'underdog status' strategy.)

This time, while the consensus among commentators is that Labor will narrowly prevail, I'm fearlessly anticipating ... it could go either way. Which is not the same as predicting a hung parliament.

If Barnett wins my $4.25 wager comes in, so that's what I'm hoping for.

Will make a final prediction on Saturday, after checking out Newspoll in the Oz (and any other opinion polls).

September 3 John McCain f*cks it up - big time

A Mumble golden rule of electoral politics: The absence of negatives trumps the presence of positives.

Further to Sunday's post.

Ultra-religious folk are no more or less perfect than you or me, but they set everyone  - including themselves - impossibly high standards. That rattling you can hear is more skeletons preparing to leap out of the closet.

I reckon John McCain has been talked into doing something really silly, when there was no need. Even in America, this sort of stuff will come a cropper, surely.

[Update: As Peggy Noonan said on open-mike: "They went for this political bullshit about narratives and youthfulness and the picture". Up-update: for a squirm-inducing example of partisan commentators saying different things in private and public, see Noonan's column - containing precisely that sort of "narrative bullshit", eg 'She could become a transformative political presence' - and the postscript at the top.]

The only question, again, is whether a dud veep candidate makes a difference. In this case it might.

May soon be revisiting and over-turning my longstanding prediction. There are two (or more) sides to every electoral equation.

September 2 Three pendulums for Western Australia

Here are the three online WA pendulums I know of. If anyone else has one I'm happy to link.

In alphabetical order: Antony, Malcolm and William.

August 31 Two polls in WA

See Mr Bludger. Galaxy uses a sample of 800 and has a fine record at elections. Westpoll, on the other hand, is a bit silly.

Galaxy's 51 to 49 (in the Sunday Times) is a swing of about 3 or 3 and a half percent to the Libs from the last election, which plotted against Antony's pendulum is line-ball in the extreme.

USA: a McStake? Or do dud running mates even matter? Probably not (in answer to second question).

Stay tuned for answer to the first. (I suspect yes: McCain has made a boo-boo - but see answer to second question.)

[going off topic] Delusions of grandeur part XXXIII

Just heard Tony Abbott on Insiders state that the United States, Great Britain and Australia 'are the most important countries in the world.'

What is it with this country? [back on topic]