Essential Report Microblogging Disclaimers

Frequently I post federal polling results and calculations to the microblogging sites Twitter (X) and Bluesky (links to my accounts in sidebar); here is an example I posted today:

The standard items I include are the primary votes for all parties, any two-party preferred or Labor vs One Nation "shadow 2PP" issued by the pollster, and what I call "my conversion".  "My conversion" is a 2PP figure I get by applying my own estimate of 2025 election preference flows to the primary votes released by the pollster, and is the figure I use in my Labor vs Coalition 2PP aggregate and my Labor vs One Nation 2PP estimate, both of which can be seen on the sidebar together with a link to the methods page for them.   I have been including primary vote changes recently because of interest in whether One Nation is going up or down, but probably won't keep doing so for long.  (In fact the one shown has a typo in it, the Greens are down 1 not up 1).  

In the case of one particular pollster, Essential Report, I have to add a lot of disclaimers because of the unusual properties of that poll.  And I'm tired of typing them out so often so I have decided to put them in article form to link to in the future to save the time of doing so.  If this has the effect of causing or coinciding with some changes in some of the less credible aspects of Essential's polling, I will take that too.

So here we go:

1. Different Scale: Every other significant pollster publishes primary votes with undecided respondents in some way removed or reallocated.  However, Essential does the following: if a voter picks "Don't know" on the first voting intention question they are asked which party they are leaning towards.  If they again pick "Don't know" they are left in, creating an undecided figure that is usually 5% or 6%, so the primary votes attributed to parties and independent/other total 94-95 instead of c. 100.  

In theory this should mean that the parties all get slightly lower primary votes from Essential than on average from other pollsters.  In practice this doesn't happen because of:

2. Very Low Independent/Other readings:  Essential measures the independent/Other vote via a voting option of "Independent or other party" which appears near the bottom of the voting options list.  For whatever reason this attracts way lower numbers than other pollsters' independent/other options - numbers too low to be believable.  The average for the first half of 2026 has been 5.7% (6.0% after reallocating undecided) compared to 10.9% from all other pollsters.  This category polled 15.1% at the 2025 federal election, although it is very likely that other pollsters are right in finding a substantial slice of that has since moved to One Nation.  (I would not rule out some other pollsters having Ind/other a bit low too.  I think about half of the 5.5% minor right vote would have gone over; I am not sure how much of the independent or left minor vote has.)

3. Unusual 2PP Method:  Essential calls its Labor vs Coalition method "2PP+".  Instead of presenting a two-party figure that sums to 100 such as 51-49, Essential leaves undecided voters as undecided in the 2PP so the figure might instead be 48-46 with 6 undecided.  Essential started doing this after the 2019 election polling failure, arguing that in a close election it was necessary to stress that a slim lead was not conclusive and undecided voters could yet cause the trailing party to win.  As I pointed out at the time this reasoning placed too much weight on the apparently spurious excuse that undecided voters caused polls to be wrong in 2019.  The 2019 failure was more likely caused by bad sampling, overly simplistic weighting practices and herding or a herding-like phenomenon.

4. Aberrant Respondent Flows: Essential's 2PP+ is largely based on asking voters who support minor parties which of Labor and Coalition they would preference highest; if the respondent then says they don't know then the last-election preference split for that party is used.  This is basically a respondent preferences method but for whatever reason it produces flows to the Coalition that are stronger than the respondent flows from other pollsters (how much stronger varies) and much stronger than last-election preferences.  On average the Coalition in the 2025-8 term is doing around two points better on Essential's "2PP+" (with undecided removed) than it is on my last-election estimates.  At the same time, Morgan's respondent preferences on average have the Coalition doing one point worse than last-election preferences - something that there might actually be a reason for if One Nation are taking voting intention from the right-wing end of the Others pool.  These preference flows in Essential are so odd that they should have a careful look at their operations (in particular spreadsheet calculations and whether their panel adequately controls speeding respondents) to see if they are doing anything wrong.

5. House Effect: Even after using last-election flows for Essential to negate the impact of its weird respondent flows, Essential still has better numbers for the Coalition overall than other regular pollsters - it tends to get the Coalition's primary on the high side and the Greens on the low side compared with the average.  This doesn't mean it's necessarily wrong about these primaries but is worth keeping in mind when I post conversion figures for it that are lower for Labor than those for other regular pollsters.  For the term so far, Essential is running 1.3 points better for the Coalition on my 2PP conversions than my aggregate at the time of its releases.  Some less regular pollsters (Freshwater, Spectre and Fox&Hedgehog) may have similar properties but have only released a few polls each.  

The combined effect of aspects 4 and 5 is that Essential had the Coalition ahead in four of its five 2PP+ readings in the first half of 2026.  But only one other pollster (Fox&Hedgehog in late May) has had the Coalition ahead on a pollster-released 2PP at all (51-49), and that was also the only poll by anyone that converted to a Coalition lead even to one decimal (50.3-49.7) on my conversions.  Every other poll that released a 2PP had Labor ahead at least 51-49, and for polls without a pollster-released 2PP all converted to ALP leads, though one of these (late May DemosAU) would have rounded to 50-50 to the nearest whole number.  

I also note that Essential doesn't have a good accuracy record in recent years.  It was at the bottom of my final-poll accuracy tables for federal 2019, federal 2022 and the Voice 2023 and 8th out of 10 (ahead only of Freshwater and the once-off Ipsos) for federal 2025, which is not to say it cannot perform well in the future or should be entirely ignored.  

Hungry Lies: One Nation's Food Security Fibs

Please explain!

Yesterday I was posting some comments about how tweets on the Pauline Hanson twitter account mostly are not written by her, despite being frequently written in the first person.  Tweets signed "PH" or "-PH", the supposed signs that they are actually hers, seem to have largely dried up; after 23 in 2023 (mostly re the Voice) there were according to the advanced search function just two in 2024, none in 2025 and only two so far this year.  In the process I had a broader look at what else One Nation's social media was tweeting and came across a two-minute video for One Nation's current "Fire the Liar" campaign.  One claim, voiced in what appears to be Hanson's voice, caught my attention:

"Seven million Australians can only afford one meal per day.  In a country as prosperous as we used to be.  People are screaming out for change"

Of course, times are tough and food security is a problem for too many.  But the idea that there are seven million Australians only eating one meal a day, and doing so for purely financial reasons rather than as a chosen diet or because they are too busy working and/or parenting to eat, seems exceedingly far-fetched.  Interested to see if there was anything behind this at all, I posted the following challenge:

Many One Nation supporters on Twitter are extremely defensive if there is even a whiff of criticism of their party so I got quite a bit of abuse for this challenge.  At the printable end of these contributions were people suggesting I was loaded and overpriveleged, that I was a Labor supporter, that I did that not care about the poor, that I had never been poor myself, that I should "get out of Sandy Bay more often", and that I lived in an insular Tasmanian bubble and did not see disadvantage (ha, I'm closer to Glenorchy City than I am to Sandy Bay, and I spend far more time there at the moment).  None of these claims were even close to true and I would rather like to look some of these people in the eye and tell them the facts about these things in person, because it might surprise them, but really they do not deserve to know.  

The Foodbank report

Among those who tried to argue in defence of the claim (some honestly but most being much too confident that they had found me out), the commonest source was the Foodbank reports, which most recently held that in mid-2025 20% of households (less than in 2022 or 2023!) were experiencing severe food insecurity.  But that probably isn't quite seven million people, and crucially, unwillingly averaging only one meal per day is a far more severe level of food insecurity than the average severely food-insecure person's.  The ABS definition of severe food insecurity is:

"Generally characterised by one or more members of the household having missed meals or reduced food intake and, at the most extreme, gone at least one day without food due to a lack of money for food." (My bolding).

So a household can be severely food insecure if even one member (not necessarily all) has missed or cut down on some meals, and that might mean it has only happened now and then rather than regularly.  A person who only averaged one meal a day for economic reasons would quite likely now and then miss meals entirely, and is certainly at the extreme end of the spectrum.  The Ipsos report notes that respondents were classified according to a USDA module on food security and looking at that module makes it clear that respondents can easily tick enough boxes to qualify as severely food insecure without having ever been reduced to a single meal per day, let alone on average.

Furthermore, for many households that do qualify as severely food insecure, that status at any given time is transitory or intermittent, as shown in Fig 29.  Only 18% of severely food insecure households reported frequently being in that status for several weeks at a time.  

And it is worth adding that the Foodbank survey estimates from the Ipsos panel are for whatever reasons way higher than the ABS's for the most recent comparable year (2023) which found 8.5% of households to be moderately or severely food insecure, a fraction of Foodbank's rates.  So Foodbank may not be a reliable source.  

The Salvation Army report

Some friendlier repliers helpfully suggested that the "seven million" claim had something to do with claims about food security made in Pauline Hanson's recent National Press Club speech.  In that speech the "seven million" claim does not itself occur, but there is a lengthy passage referencing a Salvation Army report.  Here is the start of it:

"Now let’s talk about the cost of living.

The Salvation Army’s latest Red Shield Report released in May this year is not easy to read.

It reveals the plight of increasing numbers of low-income Australians who now need help just to get by.

Thousands of Australians are seeking this help every day. Organisations like the Salvos are so overwhelmed with demand they’re having to turn people away.

Of the 4,400 people surveyed:

• 19 per cent said they’d eaten food from rubbish bins in the past 12 months;

• 60 per cent said they’d eaten expired or spoiled food;

• 91 per cent said they had skipped meals and 32 per cent said they do so on a daily basis;

• 35 per cent said they survived on only one meal a day;"

But at no point does Hanson actually inform her audience of who the people surveyed were. She makes it sound like it could be an ordinary poll or perhaps a sample of low income Australians.  It is not.  The respondents were "4,400 Australians seeking emergency relief support from The Salvation Army".  This is a survey of people who are on the whole highly disadvantaged to begin with.  The Foodbank survey claimed that 25% of food insecure households (just over 8% of all households even if we accept their high estimate of insecurity) had accessed help during the past year.  And those who frequently access such help would be more likely to be in the sample for the Salvos survey than those who had done so once or twice.   So the Salvos sample is most likely on average the most food-disadvantaged few percent.   (Hanson's description of what the report found is also slightly incorrect, since it actually said that 35% of respondents "reported on most days, could only afford one meal per day".)  Even on that level, the number of Australians in this category could be, say, hundreds of thousands.  Way too many but it isn't even one million on the numbers as published, let alone seven.

Where did the "seven million" figure come from?  It's hard to tell because it seems to have not appeared anywhere else at all before the "Fire the Liar" ad.  But it seems quite plausible that some nong staffer simply took the 35% from the Salvos report and multiplied it by, say, the adult population.  Or maybe someone just handwaved and made it up.  

But their heart is in the right place ...

Oh no it isn't, and even if it was, that's no excuse.  The MO of One Nation forever has been to scare people into believing the country is going directly to hell and that only shunning an endless list of outgroups and retreating into their tedious flagtwiddling "monoculture" can save us.  (They're cherrypickers about this anyway since they want to degrade the compulsory preferential voting system that has been at the core of Australia's electoral culture for over 100 years.)  It's working particularly well for them at the moment because after COVID and then years of cost-of-living crises, voters are very discontented - the huge challenge for establishments is a general feeling that human life just isn't getting better anymore.  But I hope it is going to stop working quite so easily and move to a phase where the voters expect One Nation to get serious and demonstrate competence - because the nation has suffered far worse with a lot less of this nonsense.  I've even had One Nation supporters saying that 2026 is the worst time in our nation's history, that it is almost impossible to stay alive ... spare me.  Real concern about a problem begins with an accurate grasp of the size and nature of the problem so you know how to address it, not with a dystopian poverty porn schlock story where the problem is too large to do anything.  

And the other thing here is that anyone with the remotest grasp of social reality should hear something like seven million Australians only being able to afford to eat once per day and think "that cannot possibly be right".  While I learned more than I ever thought was out there about often dubious "one meal a day" diets in researching this article, on the whole eating one individually insufficient meal per day on a regular basis won't be good for most people.  If a quarter of Australians were really doing it, we would be constantly hearing about what the undereating epidemic was doing to our health system.  Instead we hear far more about people eating too much of the wrong stuff to the point where two thirds of Australian adults are overweight.  If a quarter of all Australians were also eating only one meal a day there would be hardly any normal eaters left!

An MP of ministerial let alone Prime Ministerial ability should be able to grasp that this claim doesn't make sense in seconds.  Pauline Hanson (if that's really her and not an AI facsimile) either doesn't realise this doesn't make sense, knows it doesn't make sense and says it anyway, or doesn't care if it makes sense because she is dependent on her staffers and can't think for herself about this stuff.  It's hard to really say which of these is worse.

Disclaimer For Extremely Foolish People:  The author does not believe that this article will cause the electoral downfall of the One Nation political party.

----------------------

PS (26 June): In relation to David Farley voting incorrectly on a division, the Hanson account has now tweeted "I'll tell you upfront, One Nation won't always be absolutely perfect.

We'll make a mistake here and there, but we'll be upfront about it." 

and also:

"For decades, this country has been driven into the ground by politicians who won't admit they've done anything wrong, and still won't admit they're currently doing wrong.

It's time to change direction."

When will the party admit they have made a mistake with this ad?  

PS 2 July: I could now list every time Pauline Hanson or One Nation bags someone else for failing to admit they were wrong, while failing to admit One Nation was wrong about seven million Australians eating one meal a day.

Today re Moira Deeming's failure to apologise to Matthew Guy, Hanson: "It tells me a person is not prepared to admit that they got it wrong.  I want a person with integrity and honesty.  And I don't see that." 

Is Pauline Hanson prepared to admit that she got it wrong re seven million Australians eating one meal a day.  Or does she lack the integrity and honesty to admit she got it wrong?

---------------------------------------------------------------------

"Independent Australia": The Worst Website In Australian Politics?

In an article way back in 2015 I noted that the faux-progressive website "Independent Australia" was independent of "Quality control, consistency, accuracy and editorial skill."  Has it got better since?  Well no, if a recent attempt to soothe reader concerns about the One Nation surge is any guide, it seems to have got even worse.  I've decided to write a whole piece about this trainwreck to explain how abysmal IA's standards are and why nobody should be enabling their output in any way.

The most recent case is an article attributed - I have no idea whether correctly - to a Dr Jason Foster of RMIT.  Even the qualifications of the claimed author are something IA cannot seem to get straight.  The RMIT website states that Foster "holds a PhD in Media and Communication from RMIT, focusing on the representations of history in film and other media."  IA however claimed his PhD covered "media representations, politics and national consciousness", a claim that has since, without acknowledgement of error, been removed.

Before I deal with what is uniquely bad about this article and IA's failed attempt (?) to fix it, I will pick off a number of the low-hanging fruit, many of which are common myths about elections.

Begging For Debunking: A Silly Article Re Stafford Preferences

It was as sure as night follows day, as sure as daylight savings fading the curtains (oh wait) that the Stafford by-election result would see the Queensland right again beating the drum for optional preferences.  On cue we have an op ed by Morgan Begg from the IPA in the Courier-Mail trying to argue that compulsory preferential voting is some kind of aberration that has saved Labor's seat with the aid of obscure leftoids.  Begg's arguments contain a remarkable number of errors, but are also typical of the misguided and poorly debated push to return to full OPV in Queensland.

Begg is probably not responsible for the headline but for starters compulsory preferential voting is hardly an "Absurd Qld voting quirk".  Rather it is the standard form of preferential voting as it exists federally and currently (with very minor differences) in four states - and as it has existed federally, continuously, for 108 years.  (OK between 1984 and 1996 we did have a federal savings provision which allowed for a voter to deliberately exhaust their preferences, but few voters knew about that). 

Pre And Post Budget Federal Poll Roundup

2PP Aggregate 52.4 to Labor vs Coalition (term low, -0.8 since last week)

Shadow-2PP 52.9 to Labor vs One Nation (-0.6 since last week)

Labor would win election"held now" but most likely with only a small majority

This is my usual annual post about federal voting intention polling after the Budget, plus a summary of what's been happening in the months leading up.

The briefest summary of what happened in the months leading up is "not much".  In terms of my Labor vs Coalition two-party preferred aggregate, Labor dipped down to the high 52s in mid-January (off not a lot of polling at the time) but got back above 54 during Coalition leadership tensions in early February.  Since Angus Taylor replaced Sussan Ley Labor's 2PP has bobbed around the 53s with no real evidence of signal.

EMRS: The State Where No Party Has Votes

EMRS Lib 25 (-4) ALP 24 (+1) ON 19 (+5) GRN 14 (-1) IND 16 (+1) others 2 (-2)

Seat estimate off this poll if election "held now" Lib 8-9 ALP 10 ON 8 Grn 4-5 IND 4 others 0

The funniest thing about this week's EMRS poll is that it was taken before.  Before we found out on Friday that, in proof of assurances that TT Line couldn't possibly be insolvent because the government could just keep throwing it money, the embattled shipping company would be flicked a lazy half a billion dollars to keep it afloat.  Or perhaps I should better say, adrift.  Before we found out, also on Friday in federal budget week, that TasInsure, a "state-backed insurance company" floated out of nowhere in the 2025 election campaign in a desperate attempt to talk about anything at all except the stadium, had gone to the great bus mall in the ground and was being refashioned as a watchdog-shaped object.  And perhaps most significantly of all, before whatever lurks in this week's state Budget.

Stafford By-Election: Prospects and Live

Stafford (ALP 5.32%)

Luke Richmond (ALP) vs Fiona Hammond (LNP) and others

Cause of by-election: death of Jimmy Sullivan (ALP/IND)

ALP retain with c. 4.4% swing to LNP

---

Monday:  Not much more to see

Casey Briggs has tweeted that he understands there are not more than (and probably less than) 2000 postals to come, so further changes to the current 51.2-48.8 margin will be in tenths of a percent if that.

Sunday: Excuses excuses...

Steven Miles has been quoted offering a bizarre excuse for his party's poor result, claiming it was due to One Nation not running and saying "We will never know what the result would have been if they had run and not directed their supporters to vote for the LNP,".  Stafford is one of One Nation's worst seats in the state, had they run I estimate they would have got about 8%.  Most of their voters would have preferenced the LNP anyway.  Care of the 2017 election we have a window on what happens when One Nation recommends preferences to the LNP in some seats and Labor in others - the flow difference was around 10-12%.  So for an open ticket, half that.  This argument if it works at all isn't worth half a percent, it might be worth a tenth of the swing if that.  In fact not all One Nation voters would have even been aware that their party recommended (not "directed") its voters to vote for the LNP, so probably even less.  And some of those who were aware would not have obeyed.

Miles has also referred to Fiona Hammond's local profile, but that was already present in the baseline since she was the candidate last time.  Indeed her time as a councillor was more recent then.  

Another excuse quoted by The Australian is “Right-of-centre voters, after the deal with One Nation, weren’t left with many alternatives in a field of nine candidates, and so we have seen a splintering of the vote amongst other left-of-centre parties.’’  But in fact there were four right wing candidates (up one from 2024) and the three minor righties between them got a 0.3% swing on the combined One Nation and Family First primary from 2024.  It is true that Labor's primary suffered from the extra competition on the left - but that does not explain half of the primary vote swing against Labor flowing through to 2PP swing as well.  

Farrer By-Election Live

Farrer (Lib vs IND 6.2%, Lib vs ALP 12.9%)

Raissa Butkowski (Lib) vs Michelle Milthorpe (IND), David Farley (ON) and others
Cause of by-election resignation of Sussan Ley (Lib)

CALLED (7:46 pm) Farley (One Nation) gain from Liberal

-------------------------

Live comments will appear here from 6 pm - once counting gets going from around 7, refresh every 10 minutes or so for latest comments.

Monday 11th: We don't quite know the flows by party yet but what we can track is the proportion of Coalition voters whose votes would need to be switched by a different HTV recommendation for the Coalition parties' decision to preference One Nation to have decided the seat. Currently that number runs at 33.6% (this is Liberal and National combined).  As noted below, while normally it's quite feasible that an HTV decision would carry that much weight, in this case I think probably not.

Although David Farley has obviously won Farrer very easily the mathematical proof that he has done so will require a distribution of preferences meaning he's a fortnight or so from taking his seat.  See here.

The Urban Myth That "Sack Dan Andrews" Was A Labor Front

Does this look like a Labor front to you?

Group ticket voting in Victoria has again been in the news a lot lately - see my latest article about whether abolishing it would assist One Nation.  With this latest discussion has come a resurgence of a longrunning online urban myth concerning the shortlived Sack Dan Andrews party (or more formally Restore Democracy: Sack Dan Andrews Party) in the 2022 Victorian election.  The myth is that this party was set up to harvest the votes of people who hated former Victorian Premier Andrews and channel these votes back to Labor.  The reality is that while there is a disputed claim that Sack Dan Andrews (SDA) was a siphoning attempt of some sort, Labor gained no benefit from it anyway, and it had nothing to do with the party.  This article explores the reality of this short-lived party's preferences and its actual impact on the election in detail.  For those on twitter I also have a shorter version of events on a thread here.