Election Date II

Bryan · Friday 10 August 2007 · 6:24 am

Centrebet has opened a book on the election date:

Date Odds Probability
Before October 6th 2007 $67.00 1.1%
October 6th 2007 $23.00 3.1%
October 13th 2007 $21.00 3.4%
October 20th 2007 $11.00 6.5%
October 27th 2007 $9.00 7.9%
November 3rd 2007 $7.50 9.5%
November 10th 2007 $4.00 17.9%
November 17th 2007 $4.00 17.9%
November 24th 2007 $4.00 17.9%
December 1st 2007 $9.00 7.9%
December 8th 2007 $15.00 4.8%
After December 8th 2007 $34.00 2.1%

Regrets, I have had few …

Crikey reported that Newspoll chief executive Martin O’Shannessy regretted writing an opinion piece for The Australian supporting the paper’s interpretation of a recent Newspoll. I wonder whether it was the Possum’s statistical demolition of O’Shannessy’s argument that hurt.

Opinionated, ignorant and prejudiced!

Bryan · Saturday 4 August 2007 · 1:02 pm

According to Brad Norington in today’s Australian, anonymous mainstream media critics think this site often confuses fact with opinion, and bases its opinion on ignorance and prejudice.

Websites with names such as Mumble (www.mumble.com.au), Oz Politics, Possums Pollytics, Poll Bludger, Psephos, the Piping Shrike, New Matilda, Righthinker and Blogspot run commentary that mainstream-media critics say often confuses fact with opinion, and base opinion on ignorance and prejudice. All the while they feed off the commercial media, particularly The Australian.

I won’t comment on the journalistic standards that allowed such a sweeping generalisation.

For the record, I aim to keep this site analytical and non-partisan, and I try not to write anything based on ignorance and prejudice. Please contact me if you think I fail this standard.

Update: Thank you everyone who sent me emails of support.

Update #2: Larvatus Prodeo and Road to Surfdom have had something to say on this matter.

Poll wars

Bryan · Thursday 12 July 2007 · 1:32 am

According to Mumble,

A courtesy call from [the Australian's] Editor-in-Chief Chris Mitchell this morning informed me that the paper is going to "go" Charles Richardson (from Crikey) and me tomorrow.

Chris said by all means criticise the paper, but my "personal" attacks on Dennis had gone too far, and the paper will now go me "personally".

No, I’m not making this up.

If they only get as personal as I get with Dennis, then it should be tame, as I don’t believe I’ve never criticised anything other than his writing.

And to think I described Dennis, in a chapter in a book being launched this month, as (with no sarcasm) "a fine journalist".

All very strange. And - I’d be lying if I didn’t admit - a little stomach-churning.

So what did the bucket from the Australian look like? Well, it was the entire editorial in today’s Oz; the sharp end of which was the last three paragraphs.

On almost every issue it is difficult not to conclude that most of the electronic offerings that feed off the work of The Australian to create their own content are a waste of time. They contribute only defamatory comments and politically coloured analysis. Unlike Crikey, we understand Newspoll because we own it. Martin O’Shannessy understands Newspoll because he runs it and Sol Lebovic understands Newspoll because he started it. The results of our analysis speak for themselves over 20 years.

A guide book recently published by one site demonstrates the extent of confused thinking on how the polls operate. A chapter by Mumble’s Peter Brent says two party preferred ratings are at the same time worthy but unreliable and that an Opposition Leader with a high satisfaction rating has no better chance of being elected than one with a low rating. He dismisses approval ratings and the preferred Prime Minister measure as “embroidery”. Yet the fact is when Mr Howard and Mr Rudd’s offices telephone The Australian to get advance warning on what the following day’s Newspoll will show they invariably want to know two things: The primary vote and preferred PM.

Not properly understanding how polls work gives our critics licence to project their own bias onto analysis of our reporting. The Australian is not beholden to any one side of politics and recent election outcomes vindicate our treatment of our polls. So let’s not mince words. e just don’t think many of our critics have any real clue about polling and very little practical experience of politics.

Is it me or does this seem just a touch too precious?

Update: also discussed by the Poll Bludger, Surfdom, Jackman, LP, HO, Quiggan, Tug Boat Potemkin, LP #2, the orstrahyun, PO, R’n'V, Webdiary, 10,000 mile view

Update #2: Nice to see Crikey did not respond.

Update #3: It appears that the comment Australian Blogger Tim Dunlop made on this poll war is no longer visible on the Australian website. A copy can be found here.

Update #4: Confirmed, the Australian pulled Tim’s post.

Martin O’Shannessy …

Bryan · Wednesday 11 July 2007 · 7:07 am

… is the CEO of Newspoll. Today he argued that Howard’s improvement in the beauty contest polling could well be followed by an improvement in the voting intention polling.

In John Howard’s first tilt at re-election in 1998, his better PM ranking rose from 31 per cent three months out to 42 per cent just prior to polling day. In 2001, his better PM rating rose from 42 per cent three months before the poll to 52 per cent just prior to election day. In 2004 his better PM rating rose from 48per cent to 51 per cent in the three months before the poll.

Close to the election, things get hairy and the campaign runs out on the basis of whatever lead has been established in the three months before.

However, the basic pattern of improved leader’s ratings being followed by an improved primary vote appears supportable.

As we approach this year’s election, a similar pattern seems to have emerged. Mr Howard’s personal ratings dropped shortly after Mr Rudd became Opposition Leader. A close look at the past month or two suggests that the slide in the Coalition primary vote has halted.

Based on experience, it seems likely that, if Mr Howard’s personal ratings improve, a rebound in the Coalition primary vote will follow.

Update: Possums Pollytics refutes O’Shannessy (quite convincingly).