Newspoll: 54-46 to Labor; Ipsos: Labor 50, Coalition 35, undecided 15

A tick in Labor’s favour from the latest Newspoll, along with a more decisive turn in the second Ipsos poll for the campaign.

The Australian reports the weekly campaign Newspoll has Labor’s two-party lead increasing from 53-47 to 54-46, their primary vote up a point to 39% with the Coalition down one to 35% and the three minor parties steady, the Greens at 11%, One Nation at 5% and the United Australia Party at 4%. Scott Morrison’s personal ratings are deteriorated, his approval down three to 41% and disapproval up four to 55%, while Anthony Albanese is up a point to 41% and down two to 47%. Morrison’s lead as preferred prime minister has narrowed from 45-39 to 44-42.

The poll also found Labor leading 44% to 41% as best party to handle cost-of-living pressures. On this question at least, breakdowns are apparently offered by gender (44% each among men, but 45% to 38% in favour of Labor among women) and age (dramatically more favourable to the Coalition among the old than the young, as usual). The poll was conducted Wednesday to Saturday from a sample of 1523.

Also out today in the Financial Review was an Ipsos poll suggesting Labor is headed for a landslide win, with primary votes of Labor 35% (up one since a fortnight ago), Coalition 29% (down three), Greens 12% (steady), One Nation 4% (steady), United Australia Party 3% (steady) others 9% (up two) and 7% undecided (down one).

The poll offers two interpretations of two-party preferred, one of which asks respondents who support minor parties or independents to either state a preference between the Coalition and Labor or remain uncommitted, which has Labor on 50% and the Coalition on 35%, with the remaining 15% being those uncommitted on either the primary vote or the preferences question. The other allocates distributes minor party and independent preferences as per the 2019 election result, which has Labor on 52% and the Coalition on 40% with 8% identified as undecided. The accompanying report notes this translates into a 57-43 lead for Labor if the undecided are excluded.

The poll also finds 33% rate the global economy the factor most responsible for last week’s increase in interest rates, with the government on 16%, the pandemic on 17%, the Reserve Bank on 16% and the war in Ukraine on 7%. Personal ratings find Scott Morrison down two on approval to 32% and up three on disapproval to 51%, with Anthony Albanese down one to 30% and up one to 36%. Albanese’s lead as preferred prime minister widens from 40-38 to 41-36. The poll was conducted Wednesday to Saturday from a sample of 2311.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

1,964 comments on “Newspoll: 54-46 to Labor; Ipsos: Labor 50, Coalition 35, undecided 15”

Comments Page 40 of 40
1 39 40
  1. When we voted in the NT statehood referendum in 1998 I think the NT was promised an initial 3 senate seats and then a further gradual increase to a point in keeping with the ratio to Tasmania’s population. I think then as now it was about half so we were looking at gradually getting up to six senators over a series of elections.

    Which would have really frustrated people in the ACT!

  2. I just looked at the candidates in my division of Swan. Ten candidates/parties and the majority appear to be right wing nutters. The only question is if I should classify Animal Justice Party* as left wing nutters? My policy is to put the nutters last, so numbers 4-10 are going to take some work.

    There are 22 groupings for the Senate.

    I see the ALP and Liberals each are running four candidates for the Senate, with The Greens running six. I thought the two majors used to run six as well. Is there a reason they’re only running four each, and if so, why didn’t The Greens get the memo?

    * I didn’t know “netball umpire” was an occupation/

  3. I will have my two bobs worth as to the likely election outcome.

    I think the adage that governments usually lose elections is right. So why would this government lose?

    Simple, they are corrupt, they are incompetent, Morrison lies, and they have achieved nothing of value for the massive trillion dolar debt they have created.

    The opinion polls are right. It will be a wipe out.

  4. At this point I can’t see how the govt is going to be re-elected, Morrison is one of the most personally disliked people to be PM and has no vision for the country whatsoever. The election campaign has only highlighted that so it’s likely to get worse for the LNP. It will probably end up being a “surprising” landslide victory to Labor, and even in the best case scenario for the LNP there doesn’t seem to be a realistic path to a majority government. I know people are scarred by 2019 but there is nothing to suggest a second miracle is even remotely likely.

  5. The polls suggest Labor will get a majority. Albo has to keep talking about child care, medicare, aged care, renewables, TAFE, housing and a cabinet including Wong, Tanya P, Chalmers, Burke, Shorten etc.
    Scummo has to defend a trillion in debt, deficits for decades, expensive stuffed up defence contracts, inflation, increasing interest rates,family lounging at Kirrabilli, helped by Tudge, Robert, Ley, Cash, Taylor, Barnaby etc. Josh who?

  6. The key phrase to which I object there is ‘within a state’. Tasmania has an average of 80k electors per seat. The 5 big states average over 100k. A voter in Tasmania has at least 25% (in some cases, more than 40%) more say in the House election outcome than a voter in the 5 big states.

    Because somewhere in the constitution it says all of the original states need to have at least 5 members, the only way to fix this is to make the house of representatives bigger

  7. shorten was a better leader had a vishion and clear policies where as albanese is small target but morrison does not seem to be triying his heart is not in winning so labor should win based on its time factor

  8. Hazza

    “Bluepill @ 10:49pm

    You’re making a fair bit of sense there, and I don’t like it. Pretty much what Kos Samaras has been saying.

    Sorry that many on here choose to attack you when you urge caution.”

    Thanks for being reasonable, it is a rare commodity in here! One learns to have thick skin (which is why I only visit once every three years).

    Of course, everyone here thinks that I am somehow associated with the Liberal party otherwise my contributions don’t make sense. Instead, I am more interested in elections. Sure I tend to vote to the right of centre (but not always), but I am more interested in the numbers.

    Last time, the ALP has more seats with greater depth in terms of Mackerras. Last time it was a narrow LNP win but there was a (largely) unrecognised tide that went out, but the ALP managed to hang on to a bunch. However, this is no ‘Ruddslide’. Love him or hate him (and many here did in the Gillard era did, until they ‘changed their mind again’ when Rudd was leader again) he was great in front of the Mic, just like Keating was.

    Albo is apparently great in person, one on one, but is not good off the cuff and, by virtue of the last couple of years of crisis, has been more irrelevant as an opposition leader but has also missed out on the usual intense scrutiny of the opposition leader. Thus he appears less prepared. I think a ‘Team approach’ would have been better for the ALP this election. They still have a lot of cabinet that have government experience. That would spread the risk, spread the confidence from the Australian public. If I was in charge of the Coalition strategy, I would have done the same. Why have a ‘presidential campaign’ if people seem not to like the ‘president’?? It doesn’t make sense to me.

    Anyway, I am very much interested in how this pans out but, despite the TPP polls, I can see, mathematically, that the Coalition still has a better chance of forming government. If a lot of those 50/50 do literally go 50/50, then with 17 seats under 3%, stacked 12 with ALP and 5 with Coalition currently, then the Coalition would come out with about 8 or so amongst this bunch… which delivers majority government.

  9. HappyEZ

    “Yeah, but I can just look that up and see that there is way more than 69 or 71. Like, I can just look it up.
    Is that what trolling is: just lie about obvious stuff?”

    Ah… yeah mate, that is not ‘analysis’ (eye roll). I am most certainly not making stuff up. Are you analysing all markets for these seats and running confidence intervals on weighted implied probabilities? If not, then what you have is wishful thinking.

    You also seem oblivious to the fact that the ALP ideally needs 77 seats (76 +1), just like the LNP does. 69-71 on straight probability doesn’t cut it.

    You also need to consider sitting margins after redistributions. Many seats are very slim and there is clearly not much money in the seat pools, as they haven’t shifted in weeks. This affects the probability of success. I did the same thing in 2019 and the markets were , as they have been before, quite accurate. As I said, on election morning, I predicted 78 seats to Coalition, but a week later it settled to 77, I think.

    Sure there are some good possible grabs for the ALP that are close to even (but slightly short), however, the LNP has around the same amount (Bass and Braddon are good examples to compare with Lindsay and Gilmore). It is wishful thinking to believe that the ALP will get their close wins, whilst the LNP is denied theirs.

  10. Ok in answer to people’s questions:

    1. There there are no Coalition implied odds on seat betting markets that are greater than a Labor incumbent for that seat.
    2. Just like the last two elections, there are some major discrepancies in some implied odds, compared with the margin. In contrast, some show strong odds against, but not enough given the margin buffer. I applied a correction which weights the size of the margin, compared with the size of the implied probability difference and added a further correction for the TPP (less the apparent 2.2 mean bias in the top five polls last election, to balance it).

    I understand there are some assumptions here but I was trying to balance the market implied probability with the actual margin last election. In addition, I think it is fair that there is a definite swing of some description to the ALP, broadly 3% on bludgertrack, isn’t it?). However, the last two elections have seen the TPP get it wrong for 4 out of 5 major polls. So I applied 3%-2.2% to build in a margin for the ALP seats and against Independent and Coalition seats.

    When applied, this was the result:
    Coalition: 69
    ALP: 75
    Ind: 5
    GRN: 2
    Now, I must say that this algorithm caused massive disruption to current seats, which is what you tend to see in an actual election. Last time a lot of seats changed hands but still delivered a slim Coalition win.
    In this simulation, there were 31 of the 37 most marginal seats changing hands.

    Still, as you can see, there is still a prediction here for an ALP hung parliament, but seemingly no path for the Coalition.

    Of course there are several major assumptions here and many seats are so close that if any of these are substantially wrong then we’d see different results.

    I would say this model is the best chance for the ALP on these figures but it is still a hung parliament.

  11. @bluepill
    I think that 75-69 ALP win could be bang on. Any chance of an LNP majority is surely gone, and even a minority government is looking mighty tough.

    For me, past polls have been accurate around the world the past few years, but they just missed one important calculation- momentum. Whoever gets the momentum in the last week(S) of the campaign in terms of polling numbers generally wins.

    Generally you can add 2-3% to the TPP if momentum is on the government’s side, or 1-2% for the ‘opposition’ should they gain momentum. I say opposition as the momentum rule also stands up for events such as the Brexit vote in the UK.

    The problem for ScoMo is the same as Trump in 2020- people are already voting, so a late momentum shift may come too late- and that’s assuming he gains momentum.

  12. Quite Frankly the libs would be lucky to take one seat off Labor with a swing of 5 to 6% going labors Gilmore Constance is known well in less than a third of the seat which overlaps with his previous state seat. Kiama and Nowra are most of the seat where he needs to become known and win.

  13. As stated elsewhere I would like to see an analysis of the you gov model by Antony Green or similar.In my opinion it comes down to the accuracy of a sample and how that sample is reweighted . A sample of 125 is hard to be accurate could be 9% out based just on sample size. To claim a sample total of 18k is big and avoids this potential sample error size the extrapolate back wards to lots of 125..then claim this goes back to give even primary vote samples then give a preference result based on total flows.., wow a huge leap of Faith.

Comments Page 40 of 40
1 39 40

Leave a Reply

Your email address will not be published. Required fields are marked *