Nielsen Monday – Turnbull Washout Edition

Our monthly Nielsen of polling goodness comes in today with the ALP up 5 and the Coalition down 3 on the primaries to come in at 46/39 which washes into a Two Party Preferred of 56/44 to Labor – with the ALP up 4 from last month.

Running through the Approval Ratings and Preferred PM we get:

Everything is up for everyone except for the beauty contest of PPM where Rudd is up 8 to 64 and Truffles loses 7 over the last month to come in at 26.

There is an EMC poll out later today to round up last weeks data that we need for our weekly Pollytrack series and we’ll go through that when it’s released later today, so in the mean time I thought we might have a bit of a squiz at a real time version of Pollytrack.

Ordinarily we calculate Pollytrack on a weekly basis making it a weekly aggregation of all the polls weighted by sample size (for anyone that wants the gory details, they can read all about it’s methodology over at the Pollytrack page), but we don’t have to do it weekly – we can do it in real time and calculate it daily. Doing it daily makes it a little more volatile than our nice smooth weekly jobbie, but it gives a good visualisation of how much any given poll pushes the Pollytrack line around.

First up, this is what the Daily Pollytrack series looks like for Primary and Two Party Preferred votes.

The Turnbull bounce can be seen by the very small dip centred on the 22 September – as far as leadership bounces go, and contrary to some of the madness that has infected the minds of the MSM commentariat over this lately, that’s pretty pathetic.

The other thing we can do with this Daily Pollytrack is to look at the Primary and Two Party Preferred vote margins (which is simply the ALP vote minus the Coalition vote). The primary vote margin might seem like a bit of a silly thing to calculate, but in the ordinary political environment of Australia, the Coalition cannot win a Federal Election if they are behind the ALP in the primary vote – the exception to the rule here is the 1998 Election where One Nation was around delivering large amounts of preferences to the Coalition from the right, but without a big primary vote gatherer on the right, be it One Nation or whoever, the Coalition needs to be ahead of Labor to even be in with a chance.

If we calculate the Pollytrack vote margins, run a LOESS regression through it to get a line of best fit and then overlay the actual margins of each poll, and do it for both Primary and Two Party Preferred votes we get:

This chart shows just how wide the polling margins are between all the Australian pollsters we follow (Essential Report, Morgan face to face and Phone Poll, Newspoll and Nielsen). For instance, the last Nielsen was out on the margins of its MoE last time but has come back into the fold. Similarly, on the up side for Labor the Morgan face to face poll at the beginning of June showed the TPP vote for the ALP at 63 and was out there on its lonesome.

What becomes clear though from looking at the daily Pollytrack margins is that the Turnbull bounce came and went in a blink of the eye. I have a suspicion that the last three weeks of Nelson produced a bigger movement toward the Coalition than Turnbull has thus far delivered, and we might have a squiz at that tomorrow.

I’ve got to go and see a man about a dog, so our weekly Intrade update won’t be ready until about lunch time, and we’ll do Pollytrack with the EMC data either tonight or tomorrow depending on when I get the numbers.