Monday update

At the end of week two in the five week election campaign there was a slew of individual marginal seat polls, all of which were pretty ugly for Labor. At the start of the third week in the campaign, we have three national polls in. Again, they were pretty ugly for Labor:

• Newspoll came in at 46 to 54 in the Coalition's favour - a two point movement to the Coalition.
• Morgan came in at 48 to 52 in the Coalition's favour - a half point movement to the Coalition.
• Galaxy came in at 48 to 52 in the Coalition's favour - a one point movement to the Coalition.

If it were a normal polling week and Newspoll had moved two points in a week, at this point in the blog I would be noting that a two point movement is not unusual. Furthermore, it is far more likely to be noise than signal (and in most circumstances one would expect it to be followed quickly by a poll move in the opposite direction). Isolated moves in opinion polling data need to be corroborated, before being given any analytical weight.

However, this is not a normal polling week with one poll bouncing around within its normal margin of error. So far, all of the national polls have moved to the Coalition this week. And this movement is consistent with the numerous marginal seat polls we saw over the weekend, which also suggested a move in voting intention to the Coalition. It may not be by as much as two points, but some some movement towards the Coalition in the past week is the most likely scenario. This can be seen in the following aggregation (Note: updated chart from original post to accommodate an expanded Newspoll sample size).

At this point in the blog, it is my normal practice to remind people that I anchor the above Bayesian aggregation with the assumption that the net bias across all of the polling houses sums to zero. You will need to come to your own view about where the actual level of collective systemic bias lies for all the pollsters. At the 2010 Election (with a different set of pollsters), the population voting intention was about one percentage point more in the Coalition's favour compared with the pollster average (see here). In light of the 2010 experience, it is arguably plausible to subtract (say) half a percentage point or more from the above aggregation to adjust for the collective systemic bias across all of the polling houses. [As an aside, you will note that Simon Jackman, who seeks to anchor his Bayesian models with respect to the outcome of past elections regularly produces an aggregated poll that tracks well below the vast majority of individual poll results].

Having looked at the aggregation, let's pick through the entrails of the latest Newspoll. The recent changes in the attitudinal polling is perhaps the most interesting aspect of this latest release.

What are the punters thinking?

This morning my wife asked me why have I not blogged for a while. The short answer is that I am a little bored with the election campaign. There has not been much in the way of national opinion polls over the past week. While there has been a number of individual seat polls, these were all consistent with a sizable Coalition victory come September 7. Put simply, there was not much new to report.

To cure my boredom, I have been looking at the Centrebet individual seat odds. I have jerry-built a web-scraper and converted the results to the following table (the run was at 6.45pm on Saturday 17 August). Each row in the table represents a seat. The first column is the seat name. The second column is the state. The next three columns are the probabilities of a Labor, Coalition or Other win, expressed as a percentage (calculated from the odds and adjusted for the bookmaker's over-round). The "current" column indicates the party that currently holds the seat. The "favourite" column indicates the party the bookmaker's favour to win the seat. The final column is whether this would represent a change to the composition of the parliament at the time of its dissolution (Labor: 71; Coalition: 72; Others: 7). The table is ordered from the least probable result to the most probable result. Below the table are some observations.

	State	Labor	Coalition	Other	current	favourite	change
Moreton	QLD	45.7	49.5	4.8	Labor	Coalition	TRUE
Eden-Monaro	NSW	50.4	44.1	5.4	Labor	Labor	FALSE
Lingiari	NT	52.3	40.2	7.5	Labor	Labor	FALSE
Lilley	QLD	54.5	41.8	3.8	Labor	Labor	FALSE
Petrie	QLD	54.8	42.1	3.1	Labor	Labor	FALSE
Banks	NSW	39.9	54.9	5.2	Labor	Coalition	TRUE
Parramatta	NSW	55.5	37.4	7.1	Labor	Labor	FALSE
Brisbane	QLD	38.4	56.9	4.7	Coalition	Coalition	FALSE
Brand	WA	58.4	35.7	5.9	Labor	Labor	FALSE
Capricornia	QLD	58.5	35.8	5.8	Labor	Labor	FALSE
Lyons	TAS	58.6	35.9	5.5	Labor	Labor	FALSE
Page	NSW	58.8	36.0	5.1	Labor	Labor	FALSE
Blair	QLD	60.0	36.7	3.3	Labor	Labor	FALSE
Reid	NSW	35.1	60.6	4.3	Labor	Coalition	TRUE
Robertson	NSW	30.4	60.8	8.8	Labor	Coalition	TRUE
Kingsford Smith	NSW	61.0	34.7	4.3	Labor	Labor	FALSE
Solomon	NT	32.4	62.5	5.1	Coalition	Coalition	FALSE
La Trobe	VIC	31.9	62.6	5.5	Labor	Coalition	TRUE
Dobell	NSW	29.5	63.3	7.2	Other	Coalition	TRUE
Barton	NSW	63.7	29.7	6.6	Labor	Labor	FALSE
Greenway	NSW	32.8	64.4	2.9	Labor	Coalition	TRUE
McEwen	VIC	65.3	30.4	4.3	Labor	Labor	FALSE
Hindmarsh	SA	65.6	30.5	3.9	Labor	Labor	FALSE
Franklin	TAS	66.3	26.9	6.8	Labor	Labor	FALSE
Forde	QLD	28.4	66.6	5.0	Coalition	Coalition	FALSE
Richmond	NSW	67.0	26.0	7.1	Labor	Labor	FALSE
Longman	QLD	27.5	67.6	5.0	Coalition	Coalition	FALSE
Braddon	TAS	27.5	67.8	4.7	Labor	Coalition	TRUE
Bonner	QLD	27.6	68.0	4.3	Coalition	Coalition	FALSE
Melbourne	VIC	68.3	2.6	29.1	Other	Labor	TRUE
Dawson	QLD	25.7	68.4	5.9	Coalition	Coalition	FALSE
Hasluck	WA	27.9	68.8	3.3	Coalition	Coalition	FALSE
Herbert	QLD	25.9	68.8	5.3	Coalition	Coalition	FALSE
Flynn	QLD	24.8	68.9	6.3	Coalition	Coalition	FALSE
Chisholm	VIC	69.4	24.8	5.8	Labor	Labor	FALSE
Werriwa	NSW	70.8	25.3	3.9	Labor	Labor	FALSE
Adelaide	SA	70.9	23.1	6.0	Labor	Labor	FALSE
Swan	WA	23.8	72.3	4.0	Coalition	Coalition	FALSE
Deakin	VIC	20.2	72.7	7.1	Labor	Coalition	TRUE
McMahon	NSW	73.4	23.6	3.1	Labor	Labor	FALSE
Bass	TAS	20.5	74.0	5.5	Labor	Coalition	TRUE
Casey	VIC	18.1	74.8	7.1	Coalition	Coalition	FALSE
Lindsay	NSW	20.9	75.2	3.9	Labor	Coalition	TRUE
Newcastle	NSW	75.3	16.1	8.7	Labor	Labor	FALSE
Bruce	VIC	75.6	18.1	6.3	Labor	Labor	FALSE
Denison	TAS	19.5	4.9	75.6	Other	Other	FALSE
Leichhardt	QLD	18.1	75.6	6.3	Coalition	Coalition	FALSE
Boothby	SA	18.3	75.8	5.9	Coalition	Coalition	FALSE
Aston	VIC	18.2	75.9	5.9	Coalition	Coalition	FALSE
Canning	WA	19.6	76.1	4.3	Coalition	Coalition	FALSE
Corangamite	VIC	16.3	76.3	7.4	Labor	Coalition	TRUE
Dunkley	VIC	16.3	76.6	7.0	Coalition	Coalition	FALSE
Melbourne Ports	VIC	76.9	16.4	6.8	Labor	Labor	FALSE
Fairfax	QLD	6.6	77.1	16.2	Coalition	Coalition	FALSE
Fremantle	WA	77.6	16.6	5.8	Labor	Labor	FALSE
Gilmore	NSW	16.6	77.9	5.5	Coalition	Coalition	FALSE
Macquarie	NSW	16.7	78.2	5.1	Coalition	Coalition	FALSE
Bendigo	VIC	78.4	15.0	6.6	Labor	Labor	FALSE
Indi	VIC	4.0	78.7	17.2	Coalition	Coalition	FALSE
Throsby	NSW	78.9	14.4	6.7	Labor	Labor	FALSE
Rankin	QLD	78.9	18.0	3.1	Labor	Labor	FALSE
Fisher	QLD	12.2	79.0	8.8	Other	Coalition	TRUE
Hinkler	QLD	10.5	79.2	10.3	Coalition	Coalition	FALSE
Fowler	NSW	79.4	16.9	3.7	Labor	Labor	FALSE
McMillan	VIC	14.6	79.4	6.0	Coalition	Coalition	FALSE
Chifley	NSW	79.9	14.6	5.5	Labor	Labor	FALSE
Perth	WA	80.0	13.3	6.7	Labor	Labor	FALSE
Wakefield	SA	80.7	13.4	5.9	Labor	Labor	FALSE
Macarthur	NSW	14.8	80.8	4.4	Coalition	Coalition	FALSE
Fadden	QLD	10.0	81.4	8.7	Coalition	Coalition	FALSE
Bowman	QLD	13.5	81.4	5.1	Coalition	Coalition	FALSE
Wright	QLD	5.8	81.6	12.6	Coalition	Coalition	FALSE
Oxley	QLD	81.7	15.0	3.3	Labor	Labor	FALSE
Ryan	QLD	10.1	82.2	7.8	Coalition	Coalition	FALSE
Bennelong	NSW	13.7	82.4	3.9	Coalition	Coalition	FALSE
Stirling	WA	10.8	82.5	6.7	Coalition	Coalition	FALSE
Jagajaga	VIC	82.6	11.6	5.8	Labor	Labor	FALSE
Watson	NSW	83.0	12.7	4.4	Labor	Labor	FALSE
Charlton	NSW	83.3	11.7	5.1	Labor	Labor	FALSE
Isaacs	VIC	83.3	11.7	5.1	Labor	Labor	FALSE
Ballarat	VIC	83.9	10.3	5.9	Labor	Labor	FALSE
Cowan	WA	10.3	83.9	5.9	Coalition	Coalition	FALSE
Hunter	NSW	83.9	10.9	5.1	Labor	Labor	FALSE
Batman	VIC	83.9	5.7	10.4	Labor	Labor	FALSE
Makin	SA	84.2	11.9	4.0	Labor	Labor	FALSE
Sydney	NSW	84.3	5.7	10.1	Labor	Labor	FALSE
Hughes	NSW	11.1	84.6	4.4	Coalition	Coalition	FALSE
Blaxland	NSW	84.9	11.1	3.9	Labor	Labor	FALSE
Griffith	QLD	85.0	9.6	5.4	Labor	Labor	FALSE
Dickson	QLD	9.8	85.0	5.1	Coalition	Coalition	FALSE
Shortland	NSW	85.2	9.7	5.1	Labor	Labor	FALSE
Canberra	ACT	85.3	8.8	5.9	Labor	Labor	FALSE
Cunningham	NSW	85.4	8.7	5.9	Labor	Labor	FALSE
Kooyong	VIC	6.6	85.5	7.9	Coalition	Coalition	FALSE
Kingston	SA	85.5	9.8	4.7	Labor	Labor	FALSE
Lalor	VIC	85.7	7.2	7.1	Labor	Labor	FALSE
Kennedy	QLD	5.0	9.4	85.7	Other	Other	FALSE
Berowra	NSW	6.7	85.8	7.5	Coalition	Coalition	FALSE
Lyne	NSW	8.8	86.1	5.2	Other	Coalition	TRUE
Hume	NSW	7.3	86.3	6.4	Coalition	Coalition	FALSE
Corio	VIC	86.4	7.3	6.4	Labor	Labor	FALSE
Flinders	VIC	6.7	86.5	6.8	Coalition	Coalition	FALSE
Cook	NSW	6.7	86.5	6.8	Coalition	Coalition	FALSE
Wentworth	NSW	5.1	86.7	8.2	Coalition	Coalition	FALSE
Maribyrnong	VIC	86.8	7.3	5.9	Labor	Labor	FALSE
Sturt	SA	8.0	86.8	5.1	Coalition	Coalition	FALSE
Higgins	VIC	6.7	86.8	6.4	Coalition	Coalition	FALSE
Paterson	NSW	8.0	86.9	5.1	Coalition	Coalition	FALSE
McPherson	QLD	6.8	86.9	6.3	Coalition	Coalition	FALSE
Grey	SA	6.8	87.0	6.3	Coalition	Coalition	FALSE
Gippsland	VIC	6.8	87.0	6.2	Coalition	Coalition	FALSE
Barker	SA	6.8	87.1	6.2	Coalition	Coalition	FALSE
Calare	NSW	7.3	87.1	5.6	Coalition	Coalition	FALSE
Curtin	WA	4.2	87.2	8.6	Coalition	Coalition	FALSE
Bradfield	NSW	5.9	87.2	6.9	Coalition	Coalition	FALSE
Grayndler	NSW	87.2	5.2	7.6	Labor	Labor	FALSE
Mackellar	NSW	5.2	87.2	7.6	Coalition	Coalition	FALSE
North Sydney	NSW	5.9	87.2	6.9	Coalition	Coalition	FALSE
Goldstein	VIC	6.8	87.3	5.9	Coalition	Coalition	FALSE
Scullin	VIC	87.3	6.8	5.9	Labor	Labor	FALSE
Moore	WA	6.8	87.3	5.9	Coalition	Coalition	FALSE
Holt	VIC	87.5	7.4	5.1	Labor	Labor	FALSE
Hotham	VIC	87.5	7.4	5.1	Labor	Labor	FALSE
Parkes	NSW	8.1	87.5	4.4	Coalition	Coalition	FALSE
Forrest	WA	6.6	87.7	5.7	Coalition	Coalition	FALSE
Wide Bay	QLD	5.9	87.7	6.4	Coalition	Coalition	FALSE
Mallee	VIC	6.8	87.7	5.4	Coalition	Coalition	FALSE
Warringah	NSW	5.2	87.8	7.0	Coalition	Coalition	FALSE
Moncrieff	QLD	6.9	87.9	5.2	Coalition	Coalition	FALSE
Calwell	VIC	88.0	6.8	5.2	Labor	Labor	FALSE
Gorton	VIC	88.0	6.8	5.2	Labor	Labor	FALSE
Fraser	ACT	88.1	5.9	6.0	Labor	Labor	FALSE
New England	NSW	6.8	88.1	5.0	Other	Coalition	TRUE
Murray	VIC	6.8	88.1	5.0	Coalition	Coalition	FALSE
Gellibrand	VIC	88.3	5.2	6.4	Labor	Labor	FALSE
Menzies	VIC	5.3	88.4	6.4	Coalition	Coalition	FALSE
Cowper	NSW	7.5	88.6	3.9	Coalition	Coalition	FALSE
Mayo	SA	5.3	88.7	6.0	Coalition	Coalition	FALSE
Mitchell	NSW	6.9	88.7	4.4	Coalition	Coalition	FALSE
Tangney	WA	6.0	88.8	5.2	Coalition	Coalition	FALSE
Port Adelaide	SA	89.1	6.0	4.9	Labor	Labor	FALSE
Riverina	NSW	6.9	89.1	4.0	Coalition	Coalition	FALSE
Pearce	WA	5.2	89.2	5.6	Coalition	Coalition	FALSE
Maranoa	QLD	5.3	89.4	5.2	Coalition	Coalition	FALSE
Farrer	NSW	6.0	89.5	4.4	Coalition	Coalition	FALSE
Wills	VIC	89.9	3.5	6.6	Labor	Labor	FALSE
Groom	QLD	6.1	90.5	3.4	Coalition	Coalition	FALSE
Wannon	VIC	4.4	91.5	4.1	Coalition	Coalition	FALSE
O'Connor	WA	3.3	93.0	3.7	Coalition	Coalition	FALSE
Durack	WA	3.3	93.9	2.8	Coalition	Coalition	FALSE

The first thing to observe is that the Coalition is the favourite in 87 seats; Labor in 61 and Others (Bob Katter and Andrew Wilkie) in two seats. If the favourite got up in every seat, it would be a pretty convincing win for the Coalition.

	Coalition	Labor	Other
Seat count:	87	61	2

The Cube Law asserts that the ratio of seats a party wins at an election is approximately a cube of the ratio of votes the party won in that election. We can express this algebraically as follows (where s is the proportion of seats won by a party and v is the proportion of votes won by the party). Both s and v lie in the range from 0 to 1.

If we assume that Wilke breaks Labor and Katter breaks for the Coalition, we can apply the cube law to get an estimate of the TPP Labor vote implied by the punters. If we take the cube root of 62/88 we get 0.88982. Let's call this value k. We can then solve for v to show that the punters think the two-party preferred vote share will be: v = k / (k + 1). In this case, the punters are predicting a TPP vote share around 47.1 per cent for Labor.

In this context, let's look at the (so called) bellwether seats. Intriguingly, one of the historic bellwethers - Makin - is tipped to lose its bellwether status. The punters think it will stay with Labor. Other bellwether seats - Lindsay and Robinson - are tipped for a change at this election. On two bellwethers, the punters are undecided (itself an interesting fact). The odds in Moreton in Queensland and Eden-Monaro in NSW are pretty close to a line ball. In probability terms they are a coin-flip. Collectively the punters think they could go either way. While the punters think the over-all result is certain, they are hedging on these two bellwether seats.

The next thing to look at is the "others". As noted above, the punters think Andrew Wilkie and Bob Katter have a damned good chance of being returned. The punters  give Adam Bandt for the Greens less than a 30 per cent chance of being returned in the seat of Melbourne (paying $3.30 for the win). Clive Palmer's tilt in the seat of Fairfax has Centrebet paying $6.50 for "any other candidate". Similarly, Cathy McGowan's tilt in the seat of Indi has Centrebet paying $6.50 for an "Independent".

My final observation is with the captain's pick for Forde. The punters think that Bert van Manen will beat former premier, Peter Beattie.

Meaningless animation

I was hoping to update the charts with the Essential Poll from the weekend. However, it does not appear to be released. So, just for the fun of it, here is an animation of the charts I looked at this morning. The animation allows you see the charts overlaid. Yes, it is pretty meaningless.

Update: Newspoll 48-52 and Morgan 48.5-51.5

Today we have the third Newspoll in a row with 48-52 for the Coalition. More surprising has been the Morgan result with the Coalition ahead 48.5-51.5, calculated using preferences flows from from the 2010 Election. The Rudd restoration polls follow.

Using LOESS to take some of the noise out, the individual polling houses tell a story of a two to four week honeymoon, followed by a decline in two party preferred (TPP) voting intention for the ALP. (Note: with only two data points, I do not calculate a LOESS for Nielsen).

Collectively, the 30 day LOESS picture is as follows:

Caveats: the LOESS technique can be overly influenced by outlier data points at the end of the series. Today's endpoint is probably overly influenced by the last ReachTEL poll. It may also be overly influenced by the Morgan collapse, which at 3.5 percentage points (peak to trough) is almost double the next house at 2 percentage points.

Turning to the aggregation, we have:

Caveat: The zero line in the house effects chart is the average bias across the six polling houses I am tracking. You will need to come to your own view about where the actual level of systemic bias across the polls actually lies. At the 2010 Election (with a different set of pollsters), the population voting intention was about one percentage point more in the Coalition's favour than the pollster average (see here).

Because I have a soft spot for the grand old dames of Australian opinion polling, the Nielsen and Newspoll only (non-linear) aggregation follows ...

I will do another post tonight, after I have digested the Essential poll. (Update: perhaps Tuesday evening, as the data does not appear to have been released on Monday)

Newspoll in full

Around the web

Kevin Bonham has written a pretty good field guide to Australian opinion pollsters.

The ABC has an amusing vote compass thing.

Simon Jackman's tracking of betting markets suggests that last night's debate was not a game changer. For the three markets Simon tracks, things were essentially unchanged following the debate on the evening of 11 August.

Australian Political Betting, argues that while Labor won the debate with the pollsters, the Coalition won with the punters. To be fair, while more bookies moved in the Coalition's favour, there was not much movement. I suspect a bit of money come on following the debate, which has closed the field a little (the implied probabilities were all much closer following the debate).

Ben Raue has completed his seat-by-seat guide to the House of Representatives.

Update for ReachTEL 47-53 and Galaxy 49-51

Today's media had two polls, each with a one percentage point movement to the Coalition on the previous poll in its series. ReachTEL posted 47-53 in the Coalition's favour. Galaxy posted 49-51 in the Coalition's favour. It has been more than two weeks since we have had a 50-50 poll from any pollster. Today's 47-53 from ReachTEL was the first poll with a six percentage-point gap since the Rudd restoration. The updated aggregation follows. It suggests that Labor has been losing support over the past four weeks.

However, all eyes will be on tonight's debate. I am interested to see whether it is a game-changer or trend-confirmer.

Nielsen 48-52

The second Nielsen poll since the restoration of Prime Minister Rudd has moved two points in the Coalition's favour to estimate the two-party preferred (TPP) voting intention at 48 to 52 per cent in the Coalition's favour.

Interestingly, the Nielsen data suggests this Rudd restoration came in large part from a reduction in greens and others.

Unlike Nielsen, Newspoll has a sizable drop-off in the Coalition primary vote with the Rudd restoration. Nonetheless, Newspoll does have a reduction in the in green-other primary vote following the Rudd restoration.

Turning to the regular aggregation ... if we add this latest poll we find ...

If we restrict our aggregation to Newspoll and Nielson polls, we gt the following ...

Weekly aggregation

The first day of the 2013 election campaign saw opinion polls that were more favourable to the Coalition than Labor:

• Essential was unchanged at 49-51 in the Coalition's favour.
• Morgan had moved a point to the Coalition to be 49.5-50.5 in the Coalition's favour (using the preference flows from the 2010 election).
• Newspoll was unchanged at 48-52 in the Coalition's favour.
• ReachTEL has moved a point to the Coalition to be 48-52 in the Coalition's favour.

In aggregate, the polls are suggesting that Labor is at its lowest point since the restoration of Mr Rudd some five weeks ago.

At this point it's time for a caveat. The zero line in this chart is the average bias across the six polling houses I am tracking. You need to come to a view about where the actual level of systemic bias across the polls actually lies. At the 2010 Election (with a different set of pollsters), the population voting intention was about one percentage point more in the Coalition's favour than the pollster average (see here).

Unsurprisingly, an aggregated poll of 49.2 per cent TPP (on a universal swing basis) would see a small Coalition win.

However, as the 2010 campaign demonstrated, the opinion polls can move by at least three percentage points over a five week campaign. The 2013 election is still very much a contest.