Oz Politics Blog

I have upgraded the blog to WordPress version 2.2.

Please let me know if anything is broken as a result.

I am a little surprised. After a shocking opinion poll for Howard, the betting market has barely moved. The average probability of a Coalition win from the five bookmakers is 49.9 per cent — pretty well what it was a week ago (50.0 per cent).

The other graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.

The latest Newspoll was in today’s Australian. The national two party preferred vote was 59 per cent for Labor and 41 per cent for the Coalition. The primary vote predictions were 50 per cent for Labor, 36 per cent for the Coalition, and four per cent for the Greens. These results would bring little comfort to the government. And if the party hacks were scouring the data for a glimmer of hope, it is not there.

Labor is now tracking well above its Newspoll performance in previous election years. On this, Denis Shanahan noted, “If there’s no delayed Newspoll bounce or any sign of a Coalition recovery by mid-June, it will be getting too late to engineer the comebacks of 2001 and 2004.”

You may recall that I created some deliberately provocative graphs a weeks or so ago to argue the possibility that there may be a post March 2007 trend back to the government. At the time I conceded the analysis was bollocks, but I thought it was a constructive antidote for hubris or despair. When I plugged in the latest data point for Newspoll, that regression trend line moved against the Government. Now, only the series with two data points — and therefore the most unreliable series for this purpose — suggest the possibility of a post March 2007 linear trend resulting in a 5th Howard government.

Before you pop the Bollinger on ice or look for an unlocked window in a tall building, remember it only takes another data point within the margin of error to shift the Newspoll trend line back to almost the same as the Galaxy blue trend line. If the government got 44 in the Newspoll of 27 May, it would be back on track to victory.

I have said it a few times now, you cannot judge the Budget, or the likelihood of a government victory in November/December on the basis of a single poll result immediately after the Budget. We will need to wait until the end of June before we can say with some confidence that there was or was not a post March trend back to the government.

Having given you the graph that might give some (albeit not much) comfort to the Coalition, it is only fair that I add the graphs that would scare the be-Jesus out of them. Three of the four trend lines show no decline in Labor’s primary vote.

And Kevin Rudd’s popularity shows no sign of declining.

The usual opinion poll graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.

According to the Daily Telegraph, Galaxy conducted a poll of 995 voters last weekend. The headline prediction was, were an election held last weekend, Labor would have won 57 percent of the national two-party preferred vote.

Labor’s primary vote was 49 per cent. The Coalition’s primary vote was 39 per cent. The Greens are on 9 per cent. And independents and others got 3 per cent.

Of note:

Just days after Treasurer Peter Costello promised average workers a tax cut of $16 a week, only 16 per cent of those polled said they would be more inclined to vote for the Coalition, while 12 per cent said they were less inclined.

The vast majority of voters (67 per cent) said they would not be influenced either way by the big-spending Budget.

Asked whose industrial relations policies they preferred, 52 per cent endorsed Kevin Rudd while 35 per cent said they liked John Howard’s WorkChoices.

Another 13 per cent were uncommitted.

Both the DT and the Courier Mail have effectively declared it a no budget bounce poll. I just knew this headline going to happen; so I will conclude this post with an alternative interpretation of this Galaxy poll (just to be provocative).

Three weeks ago, the Galaxy TPP figure was 58 to 42 per cent in Labor’s favour. We both know that extrapolating any trend from these two data points is completely spurious. Nonetheless, the hypothetical exercise is instructive as it bodes caution against premature jubilation or despair with a change of government — depending on your political preferences. An average monthly 1.3 percentage point decline in Labor’s vote between now and the election would see the Coalition returned to Government. While a sustained average monthly decline of 1.3 percentage points over the next six months may be less likely than not, it is not implausible.

I enjoyed it. And clearly a few twists and turns to go.

What did others think?