The Flat Field: Another "Polls Are Wrong" Vid Gets It Wrong!

Five years ago I debunked a drippy poll denial video by the supposedly left wing West Media. Now here from Topher Field (not that he deserves that link) is the nearest right-wing equivalent to West's cavalcade of cluelessness.  In a video entitled "Preference Discipline is EVERYTHING" but advertised with images that say "THE POLLS ARE WRONG [AND WHY YOU ALREADY KNEW THAT]" Field, a glibertarian, sets out to show that polls that estimate One Nation vs Labor "2PPs" are underestimating One Nation.  The funny thing is that there are quite strong arguments that that is true.  He doesn't make them.  Within the first ninety seconds, Field has already showed he is totally out of his depth.

Field Doesn't Know How Polls Work

While talking in front of tweeted results of recent Redbridge and Morgan polls, Field says that these pollsters - and not just these pollsters but all the national and state polls - are using out of date data or guesses to calculate preference flows.  But in the case of One Nation vs Labor, neither of these pollsters does this.  Both calculate their preference flows simply by asking respondents which party they would preference higher.  This is hardly obscure information either - Morgan prominently states it uses this method in all its weekly polling reports, and reports on Accent Research's website make it very clear for Redbridge/Accent (see p. 3 on this one, where both methods have results shown for Coalition vs Labor but only respondent for One Nation vs Labor).   The funny thing is that very briefly some pollsters were attempting to use 2025 House of Reps preference flows but were dissuaded from doing so after I and others pointed out that this would make the flows to One Nation too high.

Currently five federal pollsters are releasing One Nation vs Labor federal shadow-2PPs - Morgan, Redbridge/Accent, YouGov, Spectre and Fox&Hedgehog.  All of them use respondent preferences.  Respondent preferences are actually problematic for estimating One Nation vs Labor because close to half of Coalition voters conventionally copy how to vote cards.  So if the Coalition recommends preferences to One Nation again then many of the Coalition voters who say they will preference Labor will actually end up preferencing One Nation.  This is the argument Field should be making if he wants to have a go at pollsters' One Nation vs Labor figures, but he hasn't bothered to do the most basic reading on how polls are actually working so he cannot do that.

He's therefore tilting at a strawman when he claims at around 2:20 that old preference flow data can't be valid when One Nation's primary has jumped fourfold, but even then he misses his imaginary opponent.  In the South Australian election despite the massive primary vote shifts, the flow of preferences involving the Coalition and One Nation was largely as expected based on modelling off the 2025 Senate election where One Nation polled a way lower vote.  And the reason his claim here is wrong is that while Coalition voters who used to rank Labor above One Nation could have become more inclined to put the now more popular One Nation ahead of Labor, that can easily be cancelled out by the Coalition voters who used to rank One Nation above Labor becoming straight 1 One Nation voters.  Every time a voter flips from the Coalition-ON-Labor ordering to ON-Coalition-Labor, the overall preference flow from the Coalition to One Nation vs Labor weakens.  Some people are quite obsessed with the idea that this means the Coalition to One Nation flows will actually be much weaker than implied by past Senate flows - as if only moderates are still voting Liberal - but I'm extremely unconvinced of that.  I think the two things roughly cancel out.

If Field had actually bothered to learn how polls actually do the One Nation vs Labor figures, he would have seen that a voter for some other party who had switched to putting One Nation above Labor without switching to voting 1 One Nation would have actually been picked up by the pollster based on the methods the pollsters are actually using.  And he would have realised that the fact that pollsters get relatively weak flows to One Nation - even if they are wrong in so doing because of how to vote cards - is actually fatal to his argument about why flows should have shifted.  So three minutes down, and two complete blunders by Topher.

Field Doesn't Know How Preferences Flow

From the fourth minute of the video he is going to tell us that the Secret Harbour results show that preferences are now flowing from the Liberals to One Nation in "unprecedented percentages".  This is in spite of the general psephosphere reaction to this by-election having been that the preference flows to One Nation were actually somewhat weaker than we were expecting!  (Something that may have been explained by many things including the Liberals having the donkey vote which flowed to Labor on a long ballot paper, overlapping federal MP Andrew Hastie's criticisms of One Nation, the Liberals picking up votes from Labor that then returned to Labor as preferences, the One Nation candidate being a former cocaine user and so on ... the reason doesn't matter, what does is that nobody has previously claimed the preference flow to One Nation from the Liberals was strong.) 

He claims he'll do this (while boasting about being "nerdy" - no mate, nerds don't get their numbers wrong, this is the difference between nerds and cranks) by comparing the data from Secret Harbour to data from Farrer to show that things are quickly changing to One Nation's benefit.  

After a few minutes of waffling Field claims that in Secret Harbour One Nation got about 89% of the preferences from the Liberals, Libertarians, Australian Christians and Shooters Fishers and Farmers combined.  Where did he get these numbers?  He's actually done this - and please remove any sharp objects from a radius of 20 metres before you read the next bit as you could well roll at least that far on the floor laughing - he's observed that One Nation went from 39.04% primary to 57.15% 2CP, a gain of 18.11 points.  He's observed that those parties together got 20.31% of the vote.  And he's divided one by the other!  

That would work if every other parties' preferences flowed 100% to Labor.  But they don't, and if Field had bothered letting even the slightest research get in the way of his theory he would have, like, looked at past elections to see what they do - just like he looked at Farrer.  So in Hunter 2025 Stuart Bonds (who scraped into the top two against Labor) got 48.5% of Legalise Cannabis voter preferences and 12.6% from the Greens.  But Field has set these both, as well as Animal Justice, four independents and even the fairly right wing Western Australia Party, all to a flow rate of zero!  

Suppose I set the flow rate from the Greens and Legalise Cannabis at the same as Hunter and implausibly assume everyone else flowed 70-30 to Labor, then suddenly Field's 89% preference flow from the mainly Liberal right wing sources falls to 69%.  Before too long we will see an actual preference flow from the Liberal exclusion point (which includes voters for other parties who put the Liberals above One Nation and Labor) and it will be nothing like 89% to One Nation.  [Edit: Antony Green has reported the flow from Liberal from a sample of about half the votes was in fact 70.2% - and that's a sample confined to the Secret Harbour prepoll place where One Nation did well (the sample included a small number of day votes).  On that basis it is likely that the overall flow from Liberal to One Nation was in the high 60s.]

The Problem With Poking The Bear

And oh dear.  I didn't know this was coming when I drafted the above text but eight minutes in Field has mentioned me, at length!  He starts by claiming that a psephologist is "an election forecaster and political pundit", which suggests he didn't do his work on that bit either - most of psephology is not forecasting, and all of it is the opposite of punditry.  

He then goes on to dispute my twitter bio claim to have "no party loyalties" but becomes yet another person to misunderstand it.  Amazingly it means exactly what it says: that I am not loyal to a party.  It doesn't mean that I am politically neutral, or have no views against any parties, or regard all equally.  If I meant to say "apolitical" I would have said it.  As I noted on the recent Poll Position pod, which Topher could have watched if he was so fascinated by me or had the slight interest in not making an even further twit of himself, the point of that line is just to warn people who are followers of any party that sooner or later I am going to have a go at it.  Some parties more than others of course, and currently One Nation about a dozen times a day, but you get that.  So Topher, you can take your advice about how to act while claiming to be neutral and give it to someone who actually makes that claim; I don't.  What I do claim is that I can have political views while also delivering neutral and accurate analysis of election results.  But thanks at least Topher for publicising my views, in particular on the rampant sookery of One Nation and the problem of fake patriots!

(He also suggests I treat Crikey as a reliable source.  In fact I treat all its pieces like any other media on their merits; he should see what I've said about Bernard Keane lately.)

Topher gets so fired up about his misinterpretation of three words in a microblogging bio that after a few minutes of this nonsense he declares that he has "no respect for [me] as a human being"!  Oh no I am so wounded, as if I would want the respect of anyone who is unable to read.  And then he declares I haven't done the work to understand myself, when he has so far showed that he has not done the work to understand: pollster preferences, the preference shift argument he is using, how preferences actually flow in elections, or what I mean by a term that actually means what it says.  And that's all in the first eleven minutes! To be accused of intellectual laziness by Topher Field, well ...

Now we're on to Field's next blunder.  Having blathered on about owning biases while failing to disclose that I have criticised him in the past and somehow ended up residing rent free in his head - accuses me of attempting to "control the narrative coming out of the Secret Harbour by-election" about One Nation's preference flows.  For some reason he does this by highlighting a tweet where I'm pointing out that One Nation's preference flows were actually not so bad as it looked earlier in the night, which is hardly the action of someone hellbent on claiming that their preference flows were crap.  Also if he actually bothered to read more of what I had said about the by-election, then far from using Secret Harbour to build a narrative of weak flow to One Nation, he would see I have repeatedly disputed attempts to generalise from it to predicting flows at other elections, and rejected calls to weaken flows in my own model because of it. So again, Topher cannot read, he doesn't do his research, and he doesn't care if he represents people he is criticising accurately.  

He goes on with some pseudomathematical babble about the 49% flow of preferences to One Nation being a supposed irrelevant denominator.  After several minutes of him again pretending to be a nerd with a brief break for flogging his merch, his complaint about this is that I am talking about an overall preference flow.  At present it's the only flow we have.  The fact is that overall flow to One Nation was somewhat lower than I would expect based on how the preferences of all the parties contesting have flowed in other contests.  And it's not just me who got that either - all the sources doing on the night projections were expecting the overall flow, given who the primary votes were for, to be over 50%.  In doing those models we were all already factoring in that voters for the Greens have very strong "preference discipline" to Labor.  But very strong is not 100%, or even nearly 100%, and even the difference between 100-0 and 90-10 can be a big deal in these cases.  As for the rest of the left, only Victorian Socialists nationally really display similar preference flows to the Greens.  Animal Justice were only 66% to Labor vs Coalition at the last federal election, Legalise Cannabis only 61% (and there is a lot of overflow between their support base and One Nation's, so the figure to Labor vs One Nation is probably weaker) and so on.  So Field just doesn't have a clue because, again, he hasn't done his research.  He has also chucked all the very random independents in this seat and even the WA Party in with the "left".

He gets back to Farrer where he claims the flow of 38.1% of preferences at the 3CP point from the Liberals to Michelle Milthorpe ahead of David Farley is evidence of preference indiscipline on the right.  But Farrer is a very local electorate and some people may have chosen the order Liberal-IND-ON based on which candidates they knew.  Another factor here is that the Liberals drew top spot on the ballot and the donkey vote flowed to Milthorpe.  Some Liberals may have been unconvinced that Farley was a real right winger anyway.  Candidate gender could have affected preference flows especially with Farley's past comments about Julia Gillard.  And so on.  Eventually he admits to some of these factors in one of the video's few moments of passing sanity, as well as noting that Farrer was independent vs One Nation and not a valid baseline for a Labor vs One Nation flow.  So, why did he even mention it?  I should also note on Farrer that breakdowns of preference flow by party voted for exist (and show the 1 Liberal flow to Farley was only 59% while the 1 Nationals flow was 69%); there is no need to just use the Liberal 3CP flow to talk about these things.  

26 minutes in he spells out what I'd already worked out - that he has assumed every single one of the preferences received by One Nation came from the right, although this assumption is patently ludicrous and shows him up as someone who should just shut up about election statistics forever.

Field then gets back to the Morgan poll.  He takes the 29.5 for Labor and adds the 13.5 from Greens (with their alleged 100% preference flow) and gets 42 (it's actually 43).  Then he queries where Labor gets the rest from.  And here he makes a still further blunder in thinking that Others is entirely other parties (left and right).  He ignores the fact that much of it is independents (or the ex-independents in Community Strong), whose preferences flow increasingly to Labor.  Indeed if he was being consistent with his Secret Harbour methods he would give Labor all the independent preferences, at which point Labor are getting more like three-quarters of others and even off his (wrongly low) 11% off the right Labor would get to something around 55-45 vs One Nation anyway.  Field wants to have his cake and eat it too - he wants to give Labor 8.8 points out of 11.8 for Others in Secret Harbour on a platter (a flow of at least 74.5%) but when it comes to the national polls he suddenly wants to cut Labor's flow from others down to half because he's forgotten that teals exist.   

This whole video by Field is a self-parodying shambles but it has somehow had a hundred thousand views.  I previously reviewed one of Field's marbles videos on here and while it had some serious flaws I think I gave it a fair go, and pointed out that Field at that time wasn't being angry and dumb about preferences and was at least trying to do the right thing by the so-called freedom movement.  But his latest video is angry and dumb about everything it touches and strongly recalls the crankery of the failed 2012 US Unskewed Polls movement - and such a waste when there was actually a valid point that was begging to be made against respondent preferencing.  Not one that would lift One Nation to victory at the moment, but one that could be worth, say, two points for them nationwide compared with the average of the respondent polls.

Libertarians for Fascism Unite!

One final thing I will say here.  I may not have that many libertarian bones in my body these days, but I do still have a lot more than One Nation.  And one thing I remember about this movement is that you're not supposed to hand it to the fash, or anything resembling fascism in any way.  A week after One Nation were whinging about the PM calling them dangerous, and on the same day of Topher's latest drooling about the rise of this party who he considers to be mostly freedom's friends, Cory Bernardi ripped off Trump (again) and tweeted "We will end the transgender movement."  Many within Australia's struggling libertarian movement made common cause with the alt-right and the far-right during the lockdowns.  If they are to stand true to the principles they claim they need to unmake it and fast.

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

Updates 

Antony Green has published this.

I've been alerted to a second Topher Field video with more nonsense, posted to Youtube four days ago and Facebook yesterday.  In this one he starts by claiming One Nation outperformed their polling.  There simply was no seat polling for the Secret Harbour by-election (and as for betting markets being wrong, Sportsbet was wrong, Bet 365 was right, and who cares about betting anyway).  Field also wrongly says that no-one saw the result coming; local expert Martin Drum for instance correctly predicted that One Nation would win easily.  

Otherwise the video is just the nonsense I have dealt with above plus more tryhard attempts to call himself a nerd and general guff in favour of One Nation.  He also claims that 2-4% of voters cast donkey votes from top to bottom.  The donkey vote is generally at most 0.5% these days.   It was higher before party names appeared on the ballot.   If Field was actually capable of mathematical thought he would realise his claims were inconsistent here - if 2% of voters donkey vote then that caps the Liberal to One Nation flow at a maximum of 88.4% even if every other Liberal voter preferences One Nation, which of course wouldn't happen.  

10 Sep

Judgment day has come quickly for Field (who despite his thoroughly unchristian behaviour on this matter has been recently seen quoting Bible passages on twitter) in the form of the release of the actual preference distributions.  These show that at the exclusion of the Liberal candidate, a mere 59.3% of preferences flowed to One Nation (even lower than the same stage in Farrer!) and 40.7% to Labor.  However the flow rate from 1 Liberal voters would have been higher; 20.6% of the Liberal-exclusion votes came from non-Liberal candidates.  What is interesting here is that this caps the flow from 1 Liberal voters at a mathematical maximum of 74.7% - and to even get that high one would have to assume that the voters who voted for minor right parties with preferences to Liberal all then preferenced Labor.  So Field's estimate of the flow was wrong by a bare minimum of 14 points but probably low 20s.  

The distribution also shows Field's assumption of essentially zero flow from not-clearly-right candidates being falsified by One Nation gaining 1082 preferences at the exclusions of such candidates.  That's 22.9% of their combined primary vote and 3.9% of all votes, but the actual 2CP flow from such candidates to One Nation would be higher than 22.9% because some would have flowed to right candidates and then to ON.  If Field had assumed such a flow from the not-clearly-right candidates then his 89% flow from right candidates would have instead been capped at 70%.  

Secret Harbour By-Election: Prospects And Live

Secret Harbour (ALP 11.54%)
George Tree (ALP) vs Luke Herdegen (ON), Ryan Robertson (Lib) and way too many others
Cause of by-election: resignation of Paul Papalia (ALP)

CALLED (9:34) One Nation gain from Labor

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

Commentary (Comments scrolling up with latest first - refresh for latest comments)

Sunday 10:50 am I am not sure we'll ever see a Labor vs Liberal 2PP for this seat but I estimate that the Liberals would have won it, around 52-48 (by normal preference flows it would be around 53 but I have reduced that on account of it being a normally weak seat).  This suggests Labor would have lost this seat even without the One Nation surge.  Given that the average swing in government by-elections is around 5% an implied swing of 13.5% is high, but also Labor are coming off a still high baseline from the last election, which would account for some of it.  

With results like this there is always a lot of talk about high swings, but the 2CP swing is not meaningful as it is not the same combatants.  The swing against the Labor 2CP is what I call a pseudo-swing; both the pseudo-swing and my estimate of the 2PP swing are lower than in Rockingham 2023.

Sunday 6:00 am What has happened in the prepolls is that Herdegen has actually got a preference flow more like I was expecting for the by-election as a whole, 52.8% (which is not surprising given how strong the prepolls were on primaries for One Nation), and so the result has gone back to the high 50s, 57-43 at present, that was expected before the bad preference flows came in.  Overall now One Nation has 49% of preferences.  From the primaries I would have expected them to get about 52.7% without considering the ballot order, which would close at least a little of the gap.  

Victoria 2026: The Election Where All Sorts Of Things Can Happen!

Victoria Aggregate Labor 26.0 Coalition 27.3 Green 12.8 One Nation 23.7 others 10.3
(polls are probably underestimating "others" vote)
2PP estimate 52.0 to Coalition vs ALP, shadow-2PP ALP 50.1 vs One Nation

Currently Coalition would be likely to govern in minority with One Nation support, with One Nation seat tally around the mid-teens

One Nation and Labor seat numbers are highly dependent on Coalition preferencing decisions

Both major parties are still capable of majority government with a modest swing from current primaries

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

It's been forever - largely because of the amount of work I had on in the first half of the year - since the last time I looked at Victorian state polling in the leadup to this year's election in any depth but finally a good time to do so has arrived.  What I see, at the moment, is a picture with a lot of possibilities.  Arguments can be made on current polling that the Allan Labor Government, while currently behind, is not far from surviving.  Arguments can also be made that it is not far from losing the election so badly that it may not even be the second party.  For now this is a really difficult election to model - much harder than South Australia - but perhaps by polling day things will get a lot clearer.  This one is a rather mathsy modelling article that has been rated 3/5 on the Wonk Factor scale.  

Is Group Ticket Voting Really On The Way Out In Victoria?

The last week has brought the encouraging news via both The Age and The Herald-Sun that Victorian Labor is preparing legislation to scrap Group Ticket Voting at the last possible chance before this year's election. According to the reports, the legislation will be brought in in the sitting week starting July 28, and would presumably pass both houses quickly in order to meet the VEC's August deadline.  

Let's hope this is finally it for all this nonsense!  I understand the smaller parties may be reluctant to vote for a new system under which their chances of winning any seats are greatly reduced for now thanks to the Bracks Government's entrenchment blunders, so I'll cut them slack so long as they refrain from making false claims in defence of their position.  But if any major party stands in the way or does nothing this time then there is no excuse for them not caring about democracy.  

Tas Liberal Senate Turnover As Duniam And Askew To Go

I had a piece in drafts almost ready to release for weeks about this, but perhaps just as well I didn't since further developments have rendered the draft somewhat out of date.   The last few weeks have seen major news in the leadup to the expected 2028 half-Senate election with the Tasmanian Liberal ticket to lose both its incumbents.  Incumbent since 2019 Wendy Askew was to retire at the end of the term but will now resign in coming weeks while incumbent since 2016 Jonno Duniam will quit by the end of this year.  Askew's successor will get something like 20 months and Duniam's successor about 17 in the Senate in the lead-up to the next election, assuming it is on schedule and not early. 

The preselection currently underway is for the next election but I would expect that the top two candidates will inherit the seats of Askew and Duniam unless there are twists regarding availability.  

While it may seem disastrous for the Liberals to be go in with no incumbents, in some ways it is an advantage to them compared to if just one incumbent departed now.  The party now has more flexibility in planning a 2028 ticket from scratch without, for instance, feeling pressured to choose a female candidate as the sole winner of the vacancy race as might have been the case had only Askew departed. 

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"

"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.

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.

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.

Would Scrapping Group Ticket Voting In Victoria Help One Nation?

On this website I have frequently covered Victoria's ongoing failure to repeal the use of Group Ticket Voting in state Legislative Council elections.  Victoria is now the last state that still has this system, which has been scrapped everywhere else after being gamed by preference-harvesting.  In the current cycle the Electoral Matters committee in an outstanding report recommended the scrapping of Group Ticket Voting way back in July 2024, and the government has still not responded officially to that recommendation.  The clock is ticking in terms of time for the Victorian Electoral Commission to implement the changes required to move to a different system, and the Commission has said the decision must be made by August.  After recent issues involving service delivery by state electoral commissions I suggest the sooner the better.

Last week there was reporting by the Guardian this week that one Labor MP had said current Premier Jacinta Allan "had appeared reluctant to [scrap GTV] as it would benefit One Nation."  Separately I understand that the view that scrapping GTV would benefit One Nation is also espoused by some Labor lower house MPs.  Irrespective of who actually holds that view, this article is to explore this claim.  

The Guardian's article does not say why anyone holding this concern might hold it, and in the absence of any actual claimed mechanism it is not that easy to counter.  However there are at least three well known myths about how Group Ticket Voting is supposedly bad for One Nation in the modern age.  Here they are and here is why they are wrong.

The Amazing 2026 South Australian Election: Final Lower House Results And Poll Performance

SA 2026 ALP 34 (+5) Lib 5 (-9) ON 4 (+4) IND 4 (-)
(Changes from pre-election/notional; Labor gained two seats from Liberal during 2022-6 term)

Estimated 2PP ALP 57.89 vs Liberal (+3.3)

Estimated "Shadow 2PP" ALP 58.19 vs One Nation

The 2026 South Australian lower house was remarkable in so many ways.  It makes Queensland 1998 seem almost boring by comparison, except that Queensland 1998 was there first.  Maybe all elections are going to be like this now and this soon will not seem so unusual but if that's so my colleagues and I are going to have a very busy time in the future!

All manner of curious things happened here.  Finally, someone (Lou Nicholson in Finniss) won a state or federal seat from fourth on primaries; hooray we have lived to see it.  Both majors missed the 2CP in Stuart and Mount Gambier in the first such cases since Nicklin 2001.  The Liberal Opposition missed more 2CPs (29) than they made (18) and were outpolled by One Nation (unprecedented) but are still the Opposition.  Worse than that they missed nine 3CPs as well and even managed to finish fifth in Port Adelaide and Black - Black being a seat they won at the previous election!  And so on.  It was obvious this was going to be a very messy election - a little while out I thought how on earth will we ever make a pendulum from THIS - but aspects of it were even more unique than I saw coming.

South Australia Postcount 2026: Hammond (And Ngadjuri)

Hammond (Lib 5.1%)
Robert Roylance (ON) vs Simone Bailey (ALP) vs Adrian Pederick (Lib)
Roylance expected to win

(Link to tallyboard thread)

The rural lower Murray seat of Hammond was one of those that stood out in pre-election modelling as being a seat on a relatively low Liberal vs ALP margin but nonetheless being apparently fertile ground for One Nation.  And this looks like this is how things have panned out.  As I start this article Simone Bailey (Labor) holds a thin lead with 27.5% over Robert Roylance (One Nation) on 27.0% with incumbent Adrian Pederick in third on 22.1%.  The most significant preference source is independent Airlie Keen on 10.4%.  Keen ran competitively in 2022 but there has been a large swing against her with the rise of One Nation.  The rest: the Greens 4.5% Legalise Cannabis 3.3% FF 1.9% Animal Justice 1.4% Lucas Hope (IND) 1.0% Aus Family 0.5% United Voice 0.4% Fair Go 0.1%.  Yes there is a candidate in this seat who actually at present has 27 votes, their presence on the ballot paper probably costing thousands of dollars in staffing costs.

South Australia 2026: Postcount Summary, Links Hub and Basic 2CP Contests

SEATS WON ALP 34 Lib 5 ON 4 IND 4

Seats covered

Finniss (Liberal) - Liberal vs One Nation vs IND, IND has won from fourth

Kavel (IND vacancy) - Labor vs IND vs One Nation, IND stayed in top two and won easily

Hammond (Liberal) - ON win staying in top two and beating Labor on preferences

Heysen (Liberal) - Liberal vs ALP or Green, Greens narrowly missed final two with Liberal defeating Labor (Liberal probably would have won anyway)

Light (Labor) - close but Labor win

Mackillop (notional Liberal) - One Nation narrow win over Liberal

Narungga (IND) - One Nation appears to have very narrowly won subject to recount.

Morphett (Liberal) - narrow Labor win

Ngadjuri  One Nation win.  

LEGISLATIVE COUNCIL  5 ALP 3 ON 2 Lib 1 Green

South Australia 2026: Final Polls Still Have Liberals Third

Average of final week polls ALP 37.1% Liberal 17.6 One Nation 23.3 Greens 11.2 others 10.8

If polls are accurate, Labor wins election easily making several seat gains, with Liberals and One Nation probably winning a few seats each.  Independents could win a few to several seats.

The 2026 SA election has been a weird one for polls.  There's been a very decent amount for a state election, but there were four polls tightly clustered around the start of the formal campaign, Morgan a little later, then nothing til the final days.  Also while the early polls had quite a range of figures, the last-week polls by Fox&Hedgehog, YouGov and DemosAU came out with almost identical primaries.  All have Labor on 37-38%, Liberals 17-19%, and One Nation 21-23%.  Newspoll stood out a bit more with a 40 for Labor and a miserable 16 for the Liberals.  

In my previous article I outlined a way of using the 2022 Legislative Council results to roughly look at where the Liberals and One Nation might make the final two if the polling is accurate, and assuming a swing that is two-thirds proportional and one-third uniform.  I have a few more observations on this model based on the vote share average at the head of this post.  

Tall Cories: Bernardi's Bunkum About Preferences

I was going to leave a comment about this until the next SA roundup but I feel that Cory Bernardi's nonsense about compulsory preferential voting should be dealt with a little faster than that.  Normally writing a whole article about what may well have just been a two-word misspeak might be considered mountain/molehill territory.  However Bernardi's statement has been reported uncritically by at least one media source, circulated by the party, and strangest by far, given a free pass on social media by some people who have brains.  

One Nation have also claimed off the back of Bernardi's comments that "we will make preference deals a thing of the past by giving voters the choice to distribute their preferences or not. "  But in fact optional preferencing doesn't do that.  It reduces the rate of deals as some parties choose not to recommend preferences, but many parties do still recommend preferences (which is what people often mean when they say "preference deals") and may negotiate about these decisions (actual deals).  There were several accusations about "preference deals" in the 2023 NSW state election.  

EMRS: What Happens When You Take A Mess And Then Throw In One Nation?

EMRS Lib 29 (-5) ALP 23 (-2) Green 15 (-2) IND 15 (-4) ON 14 (new) others 4 (-1)

Seat estimate off this poll if election "held now" Lib 10-13 ALP 9-10 Grn 5 ON 4 IND 4-5 SFF 0-1

EMRS have released basic details of the first Tasmanian voting intention poll to include One Nation in the readout (I will add a link to the full report when it is up).  One Nation are in the process of registering for state elections but are not registered yet.  This follows a federal poll for the state they released on Friday.  

The addition of One Nation has immediately seen them record a substantial 14%, but this is well below the 24% they recorded in the Tasmanian federal poll, with the Liberal vote in particular holding up much better.  One Nation's gains have come from across the board, but especially from a government that was already down on its election result in the previous poll, meaning that when this poll is compared with the election, most of the 14 points is at the expense of the Government which is down 11.  Labor and the Greens are not so far off their state marker, Independents may in effect be down somewhat given that Tasmanian state polls offering a generic "independent" option tend to overestimate election support for independents by about 4 points.

South Australia 2026: What Can The Right Still Win?

Polling for SA election is very lopsided with right very fragmented

Estimate if Newspoll is correct: approx 1 Liberal and 3 One Nation seats

Off YouGov: approx 4 Liberal and 1-2 One Nation seats

Polling may moderate by election day

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

I've been hoping for enough material to start substantial coverage of the South Australian election and it's finally arrived with polls by Newspoll and YouGov.  Prior to that there had been a string of extremely lopsided polls last year, and a Fox&Hedgehog poll in a similar vein early last week.  The YouGov poll pretty much replicates Fox&Hedgehog's finding that the conservative side is split in two with the Liberals bleeding greatly to One Nation while Labor enjoys a massive lead.  The Newspoll is even worse for the Liberals and could (if it happened) even wipe them out completely.  My estimate is that on average for the Newspoll voting intentions the Liberals would win one seat and One Nation about three, and for the YouGov poll that the Liberals might manage three or four and One Nation just one or two; it's possible there will be about as many (or more) independents as right-party MPs.  However there is a lot still to unfold with where One Nation support goes during the campaign and whether the Liberals can improve.  

What is suggested by the polls so far really aint supposed to happen.  The Malinauskas government is only at the end of its first term, but it is federally dragged, and not even by a first term federal government at that.  John Bannon in 1985 was the last State Premier to get a seat share swing at all in that circumstance.  But these are incredible times in polling and Peter Malinauskas is very popular (with a +40 Newspoll netsat).  At the same time we have what looks like a severe disruption if not a realignment on the right nationally and the SA Libs are a disaster zone.   One has to roll one's eyes repeatedly at news that "Liberal strategists" are hoping for a sympathy vote they don't deserve and trying to argue that a viable opposition is needed.  That worked so well for the similarly hapless outfit that was reduced to two seats in WA 2021.  

Farrer By-Election 2026

Farrer (Lib vs IND 6.2%, Lib vs ALP 12.9%)
By-Election 9 May
Cause of by-election resignation of Sussan Ley (Lib)

The rolling of Sussan Ley as leader of a floundering Liberal Party has led to a fascinating by-election for her seat of Farrer.  This guide will be updated from time to time with any polling news and with items of interest re candidates etc.  I will be providing live coverage on the night of May 9.

Farrer

Much of the electoral history of Farrer has been told in Antony Green's post here.  It has had just four MPs since 1949, three Liberals and a National.  Until very recently it had only ever been of the slightest interest when vacant.  Tim Fischer easily won the seat for the Nationals at the 1984 general election.  On his retirement in 2001 Sussan Ley gained it for the Liberals by 206 votes after preferences, the result being so close only because Labor ran and recommended preferences to the Nationals.

The seat has never been 2PP-competitive, only sneaking into technical 2PP-marginal territory by a handful of votes in 1972.  It saw its first serious independent attempt when Albury Mayor Kevin Mack ran against Ley in 2019.  This attempt was so hyped that betting agencies gave Mack a roughly even chance of winning but Ley won very easily, clearing 50% on primaries with a 60.9% 2CP.  

Until the 2025 election Ley had the longest active streak of wins on first preferences in the parliament (seven) but the challenge from Voices of Farrer and Climate200 endorsed Michelle Milthorpe severely dented Ley's primary vote and Ley finished up with only a 6.2% margin after preferences.  However, Farrer remained one of the most conservative seats in the country, ranking 11th on Coalition House of Reps 2PP down from 8th in 2022.  Some might think its 2PP being near the top of the list was down to Ley's personal vote but this is actually not true at all; on above the line Senate 2PP Farrer was in fact the Coalition's fourth best seat nationwide. 

Who's in the mix?

There are five four possibly competitive forces in Farrer.  Comments re candidates will be added when known but this is not intended as a candidate guide.   All candidates can be seen in a debate here.

* Liberals.  Contesting.  Justin Clancy (MP for Albury (Lib 16.3%)) was reported as a possible Liberal contender (setting up a scenario where he and Helen Dalton might both resign their state seats to run) but isn't running.  The candidate is Raissa Butkowski, an Albury councillor.

* Nationals.  Contesting.  The Nationals candidate is Brad Robertson, a former military commander.

* One Nation.  Contesting.  Sooner than they might have liked, One Nation faces a potential get off the pot moment.  If their national polling is still as high as at present following the change in Liberal leadership then they would be expected to poll strongly in this by-election and failure to do so would damage their momentum and raise some questions about their standing in the polls. The One Nation candidate is David Farley, a prominent agricultural businessman.  As is common with One Nation candidates he already has a problematic past comment that has resurfaced - but this one is also attracting more endorsements than most.  Oh, more trouble - he claims it was an accident.  (Three times?)

Further debate broke out with news that Farley had sought to join and run for Labor in 2022 and had also donated to them as recently as 2023, although this seems to result more than anything from a dislike of the Coalition.  One Nation supporters have been defensive about this but the fact is that the party has a woeful defection rate and their habit of slurping up other parties' rejects is an obvious contributing factor to this.  

In the final days of the campaign perceptions that Farley is going to become an independent at some point, perhaps sooner rather than later, if elected are growing.  

* Milthorpe (IND).  Contesting.  Michelle Milthorpe immediately announced she would run for Farrer again.  

Also-rans 

Interesting by-elections often attract large fields of uncompetitive micro-party and obscure independent candidates, some of whom are often paper candidates from nowhere near the electorate.  

* The Greens' candidate is Richard Hendrie. 

* Family First (v 2) immediately announced they would re-run Rebecca Scriven who polled 2.15% in 2025. I have not been able to verify from her S44 declaration whether Scriven is eligible for election or not - though she plausibly could be - as this apparently depends on her year of birth.

* Gerard Rennick People First announced they were running, to the vivid interest of nobody; they came last with 2.02% in 2025.  Their candidate, James Bonniefin, as best I can tell has never lived in the electorate.

* Sustainable Australia is running Lucas James Ellis.

* Gary Pappin is an independent and a councillor on Murray River Council.  He has experience in mining, farming and water resource management.  Pappin is a Mutthi Mutthi man who has had a significant role in management of the Mungo National Park and the Willandra Lakes Region World Heritage Area.  Pappin doesn't seem to be any kind of easily classifiable independent, I would probably have put him under group 5 (others) in my federal election article.

* Legalise Cannabis are running Aimee Lee Pearson.  It appears she is ineligible for election as her S44 checklist says that she has applied to renounce New Zealand citizenship but has not received the renunciation yet.  

* Roger Woodward is a serial independent candidate with a pretty random collection of policies who has also run in Cook and Berowra.  He does not live in the electorate (in fact he lives in Hornsby in Berowra).  

* The Shooters Fishers and Farmers candidate is Peter Sinclair.

Not Contesting

* Dalton (IND) Helen Dalton is the state MP for the district of Murray, which is the western end of Farrer.  She won the seat in 2019 as an endorsed Shooters, Fishers and Farmers candidate but running a largely independent-style campaign.  She quit said party in 2022 over water issues, and was returned as an independent in 2023 with an outright majority in a field of ten candidates.  Dalton was already canvassing a possible run for Farrer in 2028.  She would have to resign her state seat to contest.  On Mar 6 Dalton confirmed she would not contest.  

* Labor:  I did not believe Labor could have be competitive if they ran and I could not see much reason for them to risk embarrassment by doing so, so I was not surprised they decided not to bother.  Independents will miss that portion of their Labor preferences that comes from how to vote cards, but it is also possible in such a crowded field that an ALP run could have squeezed out independents and stopped them winning.

* Jordi Queiruga (IND): Announced running, has notified me he has withdrawn, website taken down.

* The unregistered Riverina State Party (which supports a new state covering the Victorian/NSW border areas) expressed interest in running but this may have been contingent on its registration being approved in time, which didn't happen.  The party submitted a list of 1646 putative members to the AEC in late October 2025 but has not been advertised yet.  The advertising phase followed by AEC assessment takes more than a month and party registration froze once the writ was issued for the by-election.  (I've heard they may have passed AEC checking prior to advertising.)

Prospects

This section will be edited where needed to update it.  As of 15 March we have a messy four-way battle where it is not obvious who the final two will be.    The AEC will have a challenging task in picking which candidate pair to use for the notional two-candidate count on the night (they may go with Liberal vs Milthorpe just because those were the final two last time, though there has been one poll suggesting One Nation could make the final two.)

Farrer is a relatively strong seat for One Nation, but not super strong.  Their Reps primary of just an average 6.6% is misleading because it would have been affected by Milthorpe and the relatively large field.  In the Senate, One Nation polled 9.96% in Farrer, compared to a NSW total of 6.06%.  If their national polling swing is real and flows through to Farrer they should be looking at a primary of at very least high 20s and probably much more (because One Nation tend to get inflated swings in strong areas) - but it's not that simple.

At the 2025 election, minor right preferences did not assist Ley vs Milthorpe, with One Nation, Family First and GRPF breaking only slightly to her and Trumpet of Patriots and Shooters Fishers and Farmers preferences breaking to Milthorpe.  This suggests a lot of the voters in Farrer while anti-Labor aren't wildly pro-Coalition either and are looking for any alternative ahead of the majors (something we also saw with strong ON flows to independents in the 2022 SA election).  I'd expect on that basis that any independents that run and One Nation are to a more than obvious degree fishing in the same pond.  Perhaps this will make it hard for One Nation to get a really high primary - but it is also a problem for the independent.  

It is possible that Milthorpe could get a large enough primary from many of the seat's non-conservative voters to lead on primaries, with the question then being how strong are the flows between the conservative parties and in what order they are knocked out.  Butkowski seems a good choice by the Liberals to attempt to cover off against Milthorpe.  

Dalton would have been a dangerous opponent on preferences had she run and made the final two, but with a splintered vote and having to sit between Milthorpe and One Nation it could have been hard for her to do so.

The South Australian election showed generally not very good performances by rural/regional independents, with One Nation polling fully as well as their polling implied.  

All kinds of modelling and calculations may be entered into as to what might occur here but in a rural electorate and with this being a by-election and not a general, I suspect a lot of it will be about candidate quality and the outcome could therefore be quite different from any modelling attempt.  

Polls etc

Single-seat polling is unreliable.

An Australia Institute seat poll (presumably a uComms poll) was reported on 9 Mar taken sometime between 5 and 9 Mar with "One Nation topped the count at 28.6 per cent of the total, followed by independent Ms Milthorpe (23.3 per cent), the Liberal Party (19.1 per cent), Labor (not running - 9 per cent), undecided voters (8.6 per cent) and the Greens (3.9 per cent), and the remaining 2.3 per cent favoured another party or candidate."  The news.com.au report omitted a feeble 5.2% for the Nationals.  While the media report said 64.1% of the non-One Nation voters would not consider voting for them at all ... which doesn't necessarily mean they won't preference them.  After reallocating Labor and undecided thanks to the full poll report (with a few minor design issues not helping) I estimate the primaries at One Nation 31.4 Milthorpe 29.4 Liberal 21.7 National 7.4 Green 5.4 other 2.7 (an underestimate as other was excluded from the undecided voter follow-up), unaccounted for 1.8.  Note that Milthorpe's support could be overestimated by her being named, so there's actually a good chance here that the final two would be One Nation vs Liberal, making it possible the Liberals could retain on Milthorpe and Greens preferences.  As with many uComms polls the 18-34 age group is curiously conservative looking in this poll.  A question in the form "who do you least want to win" gets One Nation 37.1 Milthorpe 29.5 Liberal 22.2 National 11.2. 

A second uComms commissioned by Milthorpe and taken April 9-10 has One Nation ahead - on raw primaries ON 30.9 Milthorpe 30 Lib 16.1 Nat 7.1 Greens 3.8 FF 1.2 SAP 0.1 (yay!) Others 2.9 undecided 7.9, coming out to 52.7-47.3 to One Nation by respondent preferences (which I would expect to underestimate the margin because of Coalition how to vote cards favouring One Nation).  The sample has Farley 14.6 points ahead on primaries of male voters with Milthorpe 12 points ahead among female voters.  I have not found a distribution of the undecided.  The major candidates were named.  One Nation's social media account has misleadingly published the breakdown for male voters as if it was the overall poll.

Sky News said that "leaked" (by which they mean given away) Liberal polling had One Nation winning easily but provided no further details, other than it sounds like One Nation are ahead on primaries.  

In non-poll news:

A projection tweeted by the 6 News @auspoll6 Twitter account (with One Nation very narrowly winning) was incorrectly claimed to be a poll by some accounts despite being explicitly labelled as a projection based off Newspoll and as not being a by-election prediction.  The tweet has now been deleted following criticism, including from Pyxis who conduct Newspoll.  What was not stated in the tweet was that the projection (with One Nation narrowly beating Liberal) was not even a uniform swing projection off Newspoll but was a projection of what would happen in Farrer if a general election was held now based on various (as far as I'm aware) unpublished assumptions about how the primary votes found by Newspoll would be reflected in particular seats.  There is a danger even if someone grasps that such a model is a projection of them thinking that there is one particular way to project a seat off a national poll, or even that projecting a seat off a national poll is any kind of reliable exercise.  

Other projections such as MRP outputs are of limited value too.  MRP outputs are meant to be collectively indicative of groups of seat types, and not reliable for a single seat at the best of times.  But also they are not designed for by-elections.  For a by-election what we want to see is seat-specific polling, preferably neutrally commissioned, transparent and certainly without any "aided vote" preambles.

Betting

Amusingly, seat betting has been seen.  Seat betting is not reliably predictive.  As of 16 Feb an early offering was NAT 2.00 Ind 2.99 ON 4.50 Lib 4.75 ALP 34 Greens 67.  

23 Feb: Ind 2.50 Nat 2.60 ON 3.50 Lib 7 ALP 51 Greens 101

5 Mar: Ind 1.91 Nat 3.40 ON 4.33 Lib 7 ALP 51 Greens 101

As of 13 Mar the above odds hadn't moved, a different site had Ind 1.65 ON 2.75 Nat 4.50 Lib 8.00.  On 15 Mar Ind was out to 1.75 and Nat in to 3.75 on that one.  

16 Mar: Second market now at Ind 1.90 ON 2.40 Nat 3.75 Lib 9.00. 

18 Mar: Both markets similar now: Ind 2 ON 2.60 Nat 5.10 Lib 8.00 vs Ind 1.85 ON 2.5 Nat 5.50 Lib 9.00 

24 Mar: 1.95/2.39/6.40/8.50 vs 1.80/2.30/7.50/8.00

8 April: One is offline, the second is 1.75/1.95/9/13, so the market increasingly sees this as a Milthorpe vs One Nation race.

24 April: Not sure when it finally flipped (had done so by April 20) but now ON 1.40 vs 2.80 and 1.38/3.10

30 April: 1.33/3.30/19/51.

7 May: 1.18/4.75/19/51.  Another site had 1.09 vs 6.50

DemosAU: Status Quo-ish Poll, But Is There An Elephant Outside The Room?

DemosAU: Liberal 35 Labor 23 Greens 15 IND 17 SF+F 4 others 6

Although poll finds Liberals down five on election, most likely result based off this poll would be no seat change from the 2025 election

The new quarterly Tasmanian state DemosAU poll is out (link to Pulse coverage), with the pollster joining EMRS in regularly canvassing Tasmanian state politics being a most welcome development.  It's my habit to write a separate article for every new Tasmanian poll that appears at least outside campaign season, but I don't have a huge amount to say about this one.  It's actually quite similar to the November EMRS with primaries of 35-23-15-17-10 compared with 34-25-17-19-5.  

The slight improvement of the Independent vote compared to the election is of no consequence given that Tasmanian polls tend to overestimate the Independent vote anyway.  The Liberals have dropped 5% on the primary vote, which on a uniform swing places Braddon in danger to the Greens (given they beat the Greens by 5.3 points there in 2025) but I would expect some of the losses to go to candidates whose preferences would help the Liberals more than the Greens.  As for the Labor primary, hmmm ... 23%, that's not good.  DemosAU does have some tendency in polls elsewhere to have major parties a bit lower than other polls, but at this stage compared to EMRS this is not apparent here.

Poll Roundup: What Do We Do With A Split Coalition ... Or A Rampant One Nation?

2PP Aggregate 53.5-46.5 to ALP (-1.3 since end of 2025, +0.6 in two weeks)

Shadow-2PP Trend Estimate 54.3-45.7 to ALP vs One Nation.  One Nation has made accelerating gains.

Labor would easily win an election "held now", probably losing several seats

The last few weeks have been remarkably messy ones for Australian politics- and poll-watchers.  Following the Bondi massacre the Opposition pressured the Government to recall parliament as soon as possible to pass measures in response, only to end up wedging itself when it got what it asked for, with the result that the National Party has again broken away.  So should analysts and pollsters still treat these parties as the same entity, and if we don't, what especially do we do in the case of Queensland where Liberal and National party room members run in separate seats under the Liberal National banner?  In the meantime, One Nation has exceeded the ex-Coalition's total in three of the ten polls released so far this year and tied it in two others.  An election right now would make Queensland 1998 look somewhat orderly, with all manner of messy multi-way seats and probably One Nation making fifty or more 2CPs with perhaps something like twenty wins - though this stuff is very hard to model.  So is it time for a Labor vs One Nation "two party preferred" figure as well?