Seat of the week: McEwen

The Melbourne fringe seat of McEwen has long been one of Victoria’s most keenly contested marginal seats, but the addition of the Labor stronghold of Sunbury in the latest redistribution may have put an end to that.

The most electorally significant change to result from the redistribution in Victoria relates to the electorate of McEwen, a traditionally marginal seat in Melbourne’s northern hinterland which has now been rendered fairly safe for Labor. This results from the transfusion of around 35,000 voters from rapidly growing Labor-voting suburbs around Sunbury, which are counterbalanced by the loss of outer urban areas further east (20,000 voters to Casey, 13,000 to Scullin and 4500 to Jagajaga), together with 10,000 to Indi and 7,000 to Bendigo in rural Victoria. The electorate maintains a stretch of the Hume Highway including Kilmore and Seymour, together with the urban fringe centres of Gisborne, Wallan and Whittlesea. Among the areas transferred to Indi are Kinglake and Maryville, which were devastated in the bushfires of February 2009.

McEwen was created with the expansion of parliament in 1984 and held for Labor in its first two terms by Peter Cleeland, who was unseated in 1990 by Fran Bailey as part of a statewide swing which cost Labor nine seats. Cleeland recovered the seat with a 0.7% margin in 1993, but was again defeated by Bailey in 1996. In 1998 it was one of a number of marginal seats which registered a below-par swing to Labor, a circumstance that allowed the Howard government to win re-election from a minority of the national two-party vote.

Consecutive swings to Bailey in 2001 and 2004 combined with a 1.0% redistribution to put the seat outside the marginal zone, but such was the swing to Labor in 2007 that Bailey needed every bit of her 6.4% margin to hold on. At first blush the result was the closest in any federal election since Ian Viner’s 12-vote victory in the Perth seat of Stirling in 1974: Labor challenger Rob Mitchell won by seven votes on the first count, but a recount turned that to a 12-vote margin in favour of Bailey. Labor challenged the outcome in the Federal Court, but the determinations the court made regarding individual ballot papers actually increased Bailey’s margin to 27.

Fran Bailey retired at the 2010 election, disappointing Liberals who hoped the esteem she gained during the bushfire crisis would stand her in good stead in a difficult seat. The party appeared to do well in preselecting Cameron Caine, a Kinglake police officer credited with saving several lives during the emergency, but he was swamped by a 5.3% swing. This made it second time lucky for Labor’s Rob Mitchell, who won preselection with the support of the Bill Shorten-Stephen Conroy sub-faction of the Victorian Right. Mitchell had earlier won a seat in the state upper house province of Central Highlands at the 2002 election, before being frozen out by the electoral reforms that took effect in 2006.

The preselected Liberal candidate for the next election is Ben Collier, managing director of Sunbury-based information technology consultancy Collier Pereira Services.

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.

1,337 comments on “Seat of the week: McEwen”

Comments Page 19 of 27
1 … 18 19 20 … 27

  3. GD,

    Mod Lib has supped on the Kool Aid of zealotry, toked on the drug of self aggrandisement and is about to embark on journey of preachy verbosity to invinceability and beyond.

    Prepare ye for the verbal diarroeha as each post is premised by, “I got the US election right, so you should only listen to my true word”.

  5. [There is nothing wrong in principle with using polls to predict an outcome.]
    It’s not “wrong” just not reliable. Polls tell you what is happening now and what happened in the past, not what will happen in the future.

  7. guytaur

    [
    HMV has spoken.]
    I realised there really was an HMV when,by total coincidence,around the world 100% of Rupe’s “independent” newspaper editors wrote editorials supporting the WMD invasion of Iraq.

  9. Mod Lib@897

    The idea that you can take an individual point estimate and then make comments about relative likelihoods of results around that estimate fascinates me and I would love to discuss further…..next time!!!

    Well you can if you have no other data. But if you have lots of other data then looking at lots of data together will be more accurate. The individual data point is still useful for its contribution to the trend but it doesn’t mean as much in that case.

    If I was following an election I knew nothing about, and there was one poll that showed a 52:48 lead for one candidate, and that was the only poll conducted, then I would have some reason to believe that candidate to be more likely to win than not win.

    I sometimes have a similar debate about “significant figures” with geographic data. Some scientists will shave off parts of a GPS reading on the grounds that the error of the GPS device renders that level of specification of location useless, eg if your GPS gives an output in metres and the accuracy is 10 metres, then whether that output ends in a 1 or a 6 doesn’t matter. But I think it’s worth retaining the data because xxxx1 +/- 10 is a different range of probable locations to xxxx6 +/- 10. That the ranges overlap is irrelevant to this.

  10. Gary

    [

    It’s not “wrong” just not reliable. Polls tell you what is happening now and what happened in the past, not what will happen in the future.]

    There are tables you can use to predict election results based on what the % margin is and how long to the election. The probability is based on looking at people in the same position and seeing what happened.

    With a 52-48 as currently and about a year to election, I think the odds are about 55-45.

  12. [The records – as precise as we can get them – show a clear plateau of global temperatures in the period from about 1800=1850. Then there’s a perceptible trend upwards.]

    Zoom, reliable climate records now stretch back more than 12,000, thanks to dendrochronology, luminescence- testing (its more recent ‘improvement’, rehydroxylation’ might not yet work accurately on very ancient pottery) – the last 2,000 recently added after the discovery of the earliest-known Neolithic grain-growing society with the world’s oldest known religious structure Göbekli Tepe: Eastern Turkey

    Cooling and warming cycles have been normal since well before 1800; as this dendrochronological study shows Climate Change and the Fall of the Roman Empire: A new study of tree growth rings has revealed a link between climate change and the rise and fall of European civilisations, such as the Roman Empire, as does this study of preColumbian American climate fluctuations (just one of a great many – there are so many existing stone & script records, which are now being referenced against dendrochronologcal and other records) Droughts, Floods, the Medieval Warm Period and the Rise and Fall of Civilisations in Central and South America. This last one is only the Abstract of another good international per-reviewed study 2500 Years of European Climate Variability and Human Susceptibility

    [Climate variations influenced the agricultural productivity, health risk, and conflict level of preindustrial societies. Discrimination between environmental and anthropogenic impacts on past civilizations, however, remains difficult because of the paucity of high-resolution paleoclimatic evidence. We present tree ring–based reconstructions of central European summer precipitation and temperature variability over the past 2500 years. Recent warming is unprecedented, but modern hydroclimatic variations may have at times been exceeded in magnitude and duration. Wet and warm summers occurred during periods of Roman and medieval prosperity. Increased climate variability from ~250 to 600 C.E. coincided with the demise of the western Roman Empire and the turmoil of the Migration Period. Such historical data may provide a basis for counteracting the recent political and fiscal reluctance to mitigate projected climate change.]

    IOW, Climate Change has happened often before during historical periods (for which we have records), often with devastating results. Though current unprecedented (note: “unprecedented” isn’t fully established in the article) warming may be due to other non-traditional factors (inc pollution of Earth’s life-sustaining gas layers), the results of climate change, esp Immigration “push factors” and their military expression; farming, and health, imply that only idiots would ignore global warming, even if they don’t believe it is caused entirely by man-made/ industrial pollution.

    Top scientists/ archeologists and those scientists with “paleo” at the beginning of their discipline, always “triangulate” their research with geological, paleantological, ice-core/ soil-core/ bog core etc analysis and historical and mythological records (inc images).

    The first evidence that the Little Ice Age was in retreat come from weather records after c1750 (inc the incidence of freezing in the Thames R’s salt reaches); esp in England, where Enlightenment’s influence led to archiving, and gifting of existing private archives to national archives.

    Private archives, kept by Councils, Guilds, Aristocrats & Gentry, churches and (esp after the Royal Society was founded) a growing number of middle-class/ artisan scholars, inventors and people interested in the sciences, were transferred to earlier versions of the British & other main museums & inportant libraries. So great were accumulations that many were not taken from their original packing until unpacked for digitalisation – one of the reasons we can now read (in fascimile, unless you’re a great personage or scholar) such important documents as the Agincourt Roll, so we now know the names and geographic origin of Henry V’s “Happy Few”.

    One focus of collation has been references to unusual weather events/ patterns – many references cited in recent series such as Time Team, Coast and Histories of Britain, Scotland etc. These are now being cross-referenced to other historical documents, international dendrochronological records, ice- and bog- cores, volcanic & other data.

    The time at which the Little Ice Age and its records of Thames R freezing display very unusually hot summers is early 1750s, esp c52-56/7. Not only is it well documented (inc re harvest conditions & crops), it shows up in fashion idiosyncrasies eg male shirts unbuttoned, sometimes to the waist (& hipster knee-breeches), no cravats, revers-collars (cooling necks) coat sleeves cut off above the elbows. Except for the Laki Volcanic “winter” of 1783, the temperature trend line from then is upwards; whereas, since the late Middle Ages, it had been downwards or steady.

    BTW, Watt & Boulton steam engines didn’t extend beyond Cornish tin mines until c1786, were limited (&confined to the UK) until after Waterloo (1815) but not in great use internationally before the 1848 Revolutions. If the global trend-line is static until 1800, then moves noticeably upward before 1850; there are more factors at play than Industrialisation & coal-burning engines!

    It doesn’t take a change-factor as great as a meteorite strike, or the onset of massive geothermal activity etc, to change climate, sometimes very quickly. Studies of climate change in preIndustrial societies suggest that other factors may be in play (inc sun-/sun-flare cycles). UK & other weather records show the current one to have been underway since c1750; however only a total fool would fail to support the thorough clean-up of soils, oceans, seas & rivers, as well as the atmosphere; since we do know, from the work of paleo-botanists, the deleterious effects of Carbon pollution of the atmosphere.

    But we’d be equally dill-brained if we thought carbon pollution extinction would stop droughts, rising seas and ice-melting, and see temperature dropping; because that didn’t happen in preColumbian America (or Cambodia for that matter), or Greco-Roman Era, or Era, starting in the 2nd millennium BC, of massive Aryan invasions (climate migrants) along a huge sweep of Eurasia, from France &Italy, Greece, Persia, Greater India, to Sanskrit civilisations in the Archipelagos north of Australia.

  13. GG

    [

    Perhaps you should invent a Duckworth Lewis system for Polling.]

    That’s basically what it is.

    We could use it just in case the next election is rained out.

  14. [Perhaps you should invent a Duckworth Lewis system for Polling.

    That’s basically what it is.

    We could use it just in case the next election is rained out.]
    It’s been rained ON by the MSM for quite some time!

  15. Diogenes@910

    With a 52-48 as currently and about a year to election, I think the odds are about 55-45.

    In whose favour? I’d say if it was just going on polling position an Australian federal government at 48-52, with polls trending in its favour, with about nine months to go, will usually win.

    Only I don’t think this is a normal 48-52 situation because I think the Coalition 2PP is being dragged down by their leader.

    I also do take the PvO argument (made at least twice in last 24 hours!) about it being difficult for Labor to win actual seats, but only so far. If the 2PP for Labor is above 51 on election day – and the arguments that it can’t be are unsound, especially if Abbott is still there – then the seats will look after themselves.

    Interesting looking at Newspoll 1992-3. Libs had a big lead in March 92 (one year out), which suddenly dropped to generally small leads (51s, 52s with the odd flare up higher) and stayed that way. Four months out, Labor suddenly took the lead (46-54 one poll then 54-46 the next), six weeks out the Coalition got the lead back, and then the Coalition led all through the campaign and lost at the last moment.

  17. BK – trying to remember who did an editorial cartoon depicting Malcolm Turnbull & Tony Abbott as old US west snake oil salesmen, selling their NBN policy. Turnbull was wearing a top hat, Abbott with the budgies over the old style suit. It was about a month ago. Any ideas? I think you posted it here.

  19. Leroy

    [Turnbull was wearing a top hat]

    Moir, I think, does Talc in a top hat with sometimes the top popped open with a thought bubble or steam and sometimes what looks like a fart.

  25. Mod Lib did get the 2010 election result close, BUT he did so by reasoning the ALP would lose:

    [Robertson, Macq, Gilmore, Macarthur, Bennelong & Lindsay]

    So to get this so wrong – the result is luck, not judgement.

  28. Kevin Bonham @ 915

    Peter van onselen theory doesnt make sense

    If the coalition primary gets lower then 2010 say 40-42%

    the coalition will lose seats , labor gets similar or higher then 2010 they will gain seats

  29. Kevin

    [In whose favour? I’d say if it was just going on polling position an Australian federal government at 48-52, with polls trending in its favour, with about nine months to go, will usually win.]

    Those odds are based on US Senate races. How much they can be extrapolated to Australian federal elections is an open question.

  31. This is what the pro coalition isnt getting

    the coalition must improve their primary vote to pick up seats

    which means the coalition must get to 45% or more to gain a majority government

    There is no way the coalition will get a pri8umaray vote of 45% or more under abbott

    the coalition are more likely to lose seats then the pick up seats under an abbott leadership

  33. Mod

    picking a couple of predictions you’ve made in the past and then pushing them as proof that you were right is a bit of a stretch.

    Why, your Victorian one shows how unsure you were about it – the poll said one thing, you ‘thought’ another, but you accepted the poll as correct.

    If someone makes enough predictions about an event well before it, then it’s easy to cherry pick afterwards and prove you got it right.

    Moreover, what’s implicit in your statement is “I predicted X election correctly, therefore when I predict Y it is correct.”

    Even your own posts, placed here to ‘prove’ your correctness, shows that you revised numbers as you went along and that your first guess wasn’t your last.

    Which is, of course, the correct way to approach them; to change your predictions as the polls change.

    So implicit in that is a recognition that any prediction made now, on the basis of current polling, is only that; and that, closer to the election, the polls may shift, and with it your predictions.

  34. This is what the pro coalition isnt getting

    the coalition must improve their primary vote to pick up seats

    which means the coalition must get to 45% or more to gain a majority government

    There is no way the coalition will get a primary vote of 45% or more under abbott

    the coalition are more likely to lose seats then the pick up seats under an abbott leadership

    labor being in government has the advantage of picking up seats, where the coalition has to improve

    it can not afford to have an less primary vote in 2013 then it did in 2010

  37. BK

    [Every time I see an Alan Moir rendition of Popeye Abbott I burst out laughing!]
    The image is just so right so who could blame you ?

  39. Abbott short term tactic failed in 2010

    Abbott was hoping the greens and Wilkie would had supported him, because it was the liberals preferences which gave them the seat

  40. labor would had 74 seats in 2010 to the coalition 72

    if the liberals did not try to play the short term tactics game

    which failed on them badly

  42. confessions
    [
    Bishop’s legal career under spotlight]
    Cunning Abbott strikes again. Before Mesma was given the S&G “Bad Cop” role by the mad monk who gave,or even knew, a flying fcuk about her former legal life ?

  43. labor maintains 37-39% minority government likely or a slight majority government

    Labor gets 40 % or more in primary a reasonable majority government

    Coalition 40-44% – cant see them picking up the seats to win government with Abbott

  45. the OH almost got polled by newspoll
    the phone rang at around 11am , the other OH answered the phone ,we are newspoll from sydney ,are you able to answer some questions,OH heard a voice in the background ” we’ve got our quota, OK sorry thank you and hung up

  47. the coalition needs to pick up 6 to 8 seats , which i can not see under an abbott or any other leader of the liberal party in 2013

  48. rua
    [For our mutual embarrassment the Pollbludger 2010 election prediction thread.]
    One might have thought with the sundry crikey computer issues, that is one thread that could have been lost to posterity 🙁

Comments Page 19 of 27
1 … 18 19 20 … 27

Leave a Reply

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