How Did The Super Saturday Seat Polls Go?

With final results for the Super Saturday by-elections now available, including preference flows, it is worthwhile reviewing the accuracy of the seat polling for these by-elections.  Recently I have posted a few times here about seat polling - see Is Seat Polling Utterly Useless? and Why Is Seat Polling So Inaccurate?  The preferencing results of the Longman by-election are also of special interest to the debate about Newspoll preferences, as the first test of Newspoll's changed methods at any federal election.

We can put that one to bed right away: the preference flow from One Nation to the Coalition in Longman was a massive 67.74%.  This should not be taken as a sign of quite how strongly One Nation preferences would flow nationally, since they did preference the LNP in this seat but would not necessarily do so in every seat, and since LNP preferences flowed strongly to One Nation in the area at the Queensland election even in the seat where the party preferenced Labor.  But it does show that it is entirely reasonable for pollsters to assume that 2016-election preferences can't be trusted in the case of One Nation.

Super Saturday By-Elections Live And Postcount

BRADDON: CALLED Keay (ALP) retain
LONGMAN: CALLED Lamb (ALP) retain
MAYO: CALLED Sharkie (CA) retain
FREMANTLE: CALLED Wilson (ALP) retain
PERTH: CALLED Gorman (ALP) is new MP, retaining seat
------------------------------------------------------------------------------------------------------------

Braddon Swings

Here is a graph showing the relationship between the vote for the independent Craig Garland and the 2PP swing in Braddon:

You can see maps of the Garland vote and the 2PP swing over at The Tally Room.  The Liberals did very well on 2PP swing on the West Coast, were smashed on King Island and the west end of the coast (Wynyard - Stanley area where fishing is important) and got small swings in Ulverstone and Devonport-Latrobe.

Poll Roundup: National Narrowing As Super Saturday Approaches

2PP Aggregate (2016 Preferences): 51.7% to ALP (-0.5 since last week, -1 in six weeks)
With One Nation preference adjustment 51.1% to ALP
Closest position since October 2016
Labor would probably still win election "held now" but it would be close

Even psephologists have trouble with counting to big numbers sometimes. Like two.  Normally I release a new post in this series every second Newspoll week, but two weeks ago on Newspoll Monday I had an inconvenient distraction.  By the time I'd got through that and a couple of days of work my mind was so much elsewhere that I had forgotten it was time for another Poll Roundup.

Anyway, another Newspoll week has come and gone and all the current streaks noted on my Newspoll records page have continued.  These streaks are: the Coalition for most 2PP losses in a row (now at 36), Malcolm Turnbull for most Better PM wins in a row (now 56), Bill Shorten for most negative netsats in a row for an Opposition Leader (now 69) and Malcolm Turnbull for the third-most negative netsats in a row for a PM (now 47, and he has overtaken Julia Gillard for second place for longest stay in negativeland by time.)

Is Seat Polling Utterly Useless?

Advance Summary

1. Seat polls have received bad publicity because of poor results at the 2013 and 2016 federal elections, and in some other recent elections such as the WA Darling Range state by-election.

2. Because it is clear that seat polls are not very accurate, it is common for posters on social media to dismiss them out of hand as useless or so misleading as to be worse than useless.

3. Indeed, seat polls at the 2016 federal election shows they were so inaccurate that they had greater average 2PP errors than simple models based on uniform swing and national polling.

4. However, in 2016 hybrid models combining seat polling and uniform swing would have been more accurate than either seat polling or a uniform swing model alone.

5. The correct use of publicly available seat polling seems to be not to ignore it entirely, but rather to aggregate it with other sources of information including national modelling.

6. Seat polling should be most useful for races that are difficult to predict by normal means, but seat polls may be unusually inaccurate in those races too.

7. The biggest problem with seat polling is the reporting that treats unreliable seat polls as definitive verdicts that one side or the other is "winning".  In fact they are weak indicators and need to be reported in the context of other evidence.

2018 Braddon By-Election

BRADDON (Tas, ALP 2.2%) 
By-election July 28
Justine Keay (ALP) vs Brett Whiteley (Lib), minor party candidates to be declared
Cause of by-election: Resignation caused by Section 44 ineligibility
Outlook: Historic patterns suggest Labor should retain, though seat polling has been very close and at times strong for Whiteley.

With the far-off date for the Super Saturday by-elections now announced, I've decided that's reason enough to put up a guide post for the Braddon contest.  This article will be updated up til polling day with I may do similar posts for some of the other seats, but if so that will probably be only after credible polling is conducted.  (And no, a ReachTEL seat poll of Longman asking people how they would vote in a federal election rather than a by-election isn't what I had in mind here.)

At the moment many people are talking about the by-election date and whether this is some kind of political stitch-up to harm Labor or whether it is all Labor's fault because its ineligible MPs did not resign or at least arrange to be referred sooner.  By the by-election day the parties won't still be talking about that, at least not if they have the slightest sense.

The Expected "Super Saturday" By-Elections

Today's four resignations from the House of Representatives following the Section 44 disqualification of Labor Senator Katy Gallagher is expected to trigger a day of at least five by-elections, or at least a cluster of by-elections close to each other.  The following seats are affected:

Braddon, TAS (ALP, 2.2%)
Fremantle, WA (ALP, 7.5%)
Longman, QLD (ALP, 0.8%)
Mayo, SA (Centre Alliance vs Lib, 5.0%; Lib vs ALP 5.4%)
Perth, WA (ALP, 3.3%) 

See The Tally Room for detailed histories of these seats.  Also see the Poll Bludger thread for Perth.  All seats will be contested on the old boundaries, irrespective of redistributions.

It's possible that given the strictness of the High Court's ruling, other MPs may come under pressure to resign or be referred to the High Court (note: as of Friday the media are suddenly all over the Anne Aly story, which has been known via Jeremy Gans' Twitter comments for months), though the Coalition may not be in any great hurry to hunt down any more and invite more scrutiny of its own remaining unclear cases.  The by-elections are not just a nuisance for Labor, but also for the Coalition, which must either throw resources into contesting them seriously or else chicken out and leave voters wondering what all the fuss was about.

Australia has never had a day with five federal by-elections before, so it would be quite a novelty.  Three were held on the same day in 1981 and 1984.  In 1994 four were held across three weekends following a cluster of resignations, but the resignations came on different days.  At state level, NSW has often held multiple by-elections on the same day.