politicalowl

When you see opinion poll data like that below in The Australian  this morning the inclination is to remember that with six months or so to an actual polling day that things can change. Five percentage points is near enough to the the normal error with so far to go and a 45% Coalition share can easily become 50%. That's the way most people would look at it but deviations can go down as well as up. If you don't think that's so go to Wikipedia and read about the Canadian election of 1993 . The governing Progressive Conservatives were polling at 36% early in September (down from the 43% with which they had won office in 1988) but at the election at the end of October the PCs managed just 16%. And the result? Down from 169 seats to 2. Just something for Coalition members in Australia to think about.

Hardly surprising... check back in a month for any real sense.... https://t.co/1VExhMkjEg — Chris Kenny (@chriskkenny) August 26, 2018

The latest measurements of Newspoll, Ipsos and Essential all are now showing the federal election mood as 49% for the Coalition to 51% to Labor. That is a long way short of justifying the "Labor is on the way to certain victory" mood that dominates my Twitter feed after the weekend by-elections. On the contrary it suggests that while Malcom Turnbull might not be favourite to retain the Prime Ministership he is still well and truly in the contest. If PM Turnbull really believes that company tax cuts are the right thing to do then there is no evidence in the opinion poll results to abandon that policy. Voters, maybe grudgingly, tend to prefer leaders who stick to their guns rather than abandon beliefs when backbenchers start worrying

If Australia had a presidential election, the opinion polls suggest that Malcolm Turnbull would defeat Bill Shorten easily. Newspoll this morning had the Prime Minister 19 percentage points in front of the Opposition Leader when voters were asked "Who do you think would make the better PM?" So why is it that Shorten's Labor lot have for ages now been comfortably in front of Turnbull's Coalition when it comes to voting intention? That's a question the Liberal and National Parties should be thinking about. The Owl reckons, for a start, they should be hiding this pair of vote losers:

Malcolm Turnbull backed a winner at the rugby league last night and he has reason to be a tiny bit happier about some opinion polls too. While the aficionados are concentrating on Newspoll number 30, three others out this week have the Coalition doing better. Not winning, mind you. But close enough to show that the next election contest is not yet over. Ipsos       Coalition 48 Labor 52 Essential Coalition 48 Labor 52 Morgan   Coalition 49 Labor 51

Out this morning:

The final verdict of the Owl's survey was: And the result of Newspoll itself? Coalition 47 Labor 53

And the actual result? Labor 53%, Coalition 47%. Not flash for the Turnbull led team by any means but far from brilliant by mine.

Click to enlarge What will the magic number be? An increase on the 56% figure last time will surely add extreme pressure on Malcolm Turnbull when parliament returns a week later. polls

An ABC News report on the eve of the Northcote by-election: Northcote by-election: New union-commissioned poll forecasts Labor victory in Saturday ballot Labor should be able to hold on to Northcote in a by-election, according to a new poll released on the eve of the vote in the seat under threat from the Greens. Saturday's by-election in progressive Northcote pits Labor against the Greens, who have been closing the gap on their major party rivals in the seat over the past decade. ... But new polling from ReachTEL of 850 voters— commissioned by the Construction Forestry Mining and Energy Union— shows Labor leading on a two-party preferred basis 54 per cent to 46 per cent. The primary vote shows Labor's Clare Burns on 38.8 per cent ahead of the Greens Lidia Thorpe on 34.4 per cent. The poll tested support from the Animal Justice Party (3.8 per cent), Liberal Democrats (4.7 per cent) or an independent (8.7 per cent). Based on further questions to undecided and minor parties, Re...

It was a disappointingly small response from the Owl's readers as he attempted to get the verdict of an informed crowd as to what today's Newspoll would show. Nevertheless while the quantity was small the quality was excellent. The average of those predictions received was just a tick over 55% which was the published figure in The Australian. As a believer in the wisdom of crowds we are going to persist. survey solutions

Yet another bad opinion poll for the Prime Minister and his troops will be even more jittery than ever. Another one like this the day before the House of Representatives returns on 27 November and the talk of a leadership change may well become a reality. Labor's Bill Shorten should be hoping it doesn't. Malcolm Turnbull is the best thing going for him. Another month and another Liberal Leader and we can forget about a two party vote of 56% for Labor being a real indication of future voting intention. I doubt that it is at the moment actually. Leads of 10 points just don't happen at real elections. Think back to 1972 when William McMahon was Prime Minister. He was every bit as disparaged as PM back then as Turnbull is on Twitter today but ended up only a bit over five points behind. The table below compares Labor's two party vote 15 months after the previous election with Labor's actual vote at the following five elections Labor on Newspoll 15 months after an ...

Last week the Newspoll in The Australian put Labor's share of the two party preferred vote at 54%. The Owl is intrigued about what it will be next week. What do you think? picture polls

While updating my table of The Australian's Newspoll results today (you will find the collection HERE ) it struck me once again just how irrelevant poll results are a long time out from an election. The table below compares Labor's two party vote 15 months after an election (like the one this week) with Labor's actual vote at the following election. Labor on Newspoll 15 months after an election Labor vote at next election Difference in percentage points 55 49.3 5.7 43 46.5 3.5 58 50.1 7.9 48 52.7 4.7 49 47.2 1.8

Some evidence this week, if you needed it, that Nick Xenophon's Team is more in touch with your "average" Australian than any of the other parties in the federal parliament. Research firm JWS Research set out to find what Australians regard as the sensible centre . And it discovered the answer is slightly more ‘right’ than ‘left’ and slightly more conservative than progressive. Here are the key survey questions and the results: JWS summarised the results of its full survey as: Relative to how they rate the main federal political parties, Australians view themselves as sitting somewhere in the middle, as: more left leaning and more socially progressive than the Liberal-National Coalition and One Nation  more right leaning and more socially conservative than Labor and The Greens  more right leaning but similarly moderate on social issues to the Nick Xenophon Team (and this trend is even stronger in South Australia).  Similarly, relative to key federal poli...

There was one observer who got it right after Malcolm Turnbull turned feral with that personal attack on Bill Shorten early in February. Hugh Mackay, who Michael Gordon writing in the Fairfax papers described as having been studying the national mood for decades, says Turnbull has done himself immense damage. "I think we will be able to track the gradual disintegration of Turnbull from that attack," he tells me. "It was such a weak strategy and so appallingly personal and vile that I think a lot of people are going to lose a lot of respect for him over that." Today's Newspoll suggests Hugh Mackay was on the right track.

Why forecasters failed to predict Trump’s victory : Tim Harford in London's Financial Times "The truth is that once Trump had secured the nomination, a Trump presidency was always a strong possibility. The betting markets seemed to recognise this, offering odds of three-to-one a week or so before the poll. Three-to-one shots happen all the time — or at least, about a quarter of the time. A defeat for Hillary Clinton may be far more consequential than a defeat for Manchester City and, therefore, far more shocking but it shouldn’t be any more surprising. Favourites do not always win." "... we have to keep an open mind that more than one outcome is possible. Too many people equated “Clinton is the favourite” with “Clinton will win”. That’s an obvious error, but it’s common. Even expert forecasters often treat a strong possibility as though it is a certainty. This tendency is one reason that dart-throwing chimps give the experts a run for their money. The chimps make ...

The NT News does it again. Of COURSE @TheNTNews makes its election prediction using a psychic crocodile, nothing else will do 🐊🔮 pic.twitter.com/jaOSnmq6WS — Josh Butler (@JoshButler) June 28, 2016

7 News at 6pm: The exclusive poll result which could change the course of the federal election. #7News pic.twitter.com/ikHUpWNocE — 7 News Sydney (@7NewsSydney) May 27, 2016 So what did the shock poll result actually show? Labor in front 52 to 48. And what's the normal error in polls this far from election day. Around 3.5 to 4 percentage points. (See  When Should You Start Worrying About the Polls? ) So the exclusive result is Labor probably somewhere between 48 and 56. The betting market, meanwhile, has the Coalition at $1.33 and Labor at $3.70

When Should You Start Worrying About the Polls? - The New York Times : Click to enlarge "The chart above shows how much the polling average at each point of the election cycle has differed from the final result. Each gray line represents a presidential election since 1980. The bright green line represents the average difference. At this point – 167 days before the election – a simple polling average has differed from the final result by about nine percentage points. We expect this average to become more meaningful by the week, until the national party conventions temporarily make it less so, as shown in the bump about 100 days before the election. The average difference begins to flatten about two months before the election. The day before the voting, an unadjusted polling average has been about 3.5 points off the final result." 'via Blog this' The same phenomenon can be observed in Australia. The further away from election day an opinion poll is taken th...