Possums Pollytics

Politics, elections and piffle plinking

The Nightwatchman

Posted by Possum Comitatus on March 4, 2008

 crikeylogo.jpg This was me earlier today in Crikey

Politics is a lot like cricket – and I don’t just mean the sledging and the façade of civility. The Liberal top order collapsed with the Labor onslaught of the last election, but faced with Rudd taking the new ball late in the day, rather than the Libs looking for a middle order rally with their Turnbulls and Bishops – they played it defensive and sent in the nightwatchman; Brendan Nelson strode to the wicket.

The problem here is that while Labor are playing a modern version of bodyline and pounding the Nightwatchman into a bloody pulp in the process, the batsmen next in line to occupy the leadership crease seem more than happy to keep Nelson out there, absorbing the political missiles Rudd keeps bowling and they refuse to retire him hurt until he has softened that new ball up just that little bit more – but it’s also at the expense of the score.

Preferably the nightwatchman would be softening the ball up by using his bat -a bit of defensive policy work here and there with the occasional cover drive past Wayne Swan and Peter Garrett to the boundary. But after today’s Newspoll, it looks like Nelson’s been padding up to the bouncers without his helmet on.

The headline two party preferred result comes in at a thumping 63/37 to Labor off the back of primaries running 51/31 the same way. As with the previous Newspoll, it is the Preferred Prime Minister rating that really does the damage with Nelson slipping a statistically significant 2% points down to a new record low of 7%.

While Newspoll states that the maximum sampling error in this poll is 3%, when you get results that are far removed from a 50/50 poll split, the sampling error reduces substantially, to the point where the margin of error on Nelsons preferred PM rating is around the 1.5% mark.

With preferred PM ratings this low, it’s pretty clear that Nelson’s leadership is becoming a limiting factor in the Coalition’s level of electoral support – his only saving grace is that the number of uncommitted responses in the survey are still quite high, with 20% being uncommitted on preferred PM and 33% still unable to decide whether they are satisfied or dissatisfied with his performance.

One expects to see satisfaction and performance dynamics like this when the Opposition leader is failing to cut through and be noticed (think Simon Crean), but with the Coalition primary vote being so disastrous, where 1 in 4 Coalition voters at the last election have now abandoned the conservatives, it also has the whiff about it that when Nelson does cut through – the electorate doesn’t particularly like what it sees.

This doesn’t look like a sustainable political position for the Coalition to be in with a bag full of by-elections coming up. But would the Liberal Party risk exposing their other leadership contenders too early to the Labor onslaught by pulling the plug on Nelson soon – or are the leadership aspirants just content that it’s the hapless nightwatchman and not they whom are taking the shine off Labor’s new ball?

The possible upcoming by-elections of Mayo, Higgins, Gippsland and Lyne could well be decided by what happens next.

15 Responses to “The Nightwatchman”

  1. The Keegan said

    I see the Nelson nickname competition has settled on ‘Nightwatchman’ – perfect! Your analysis of Lyne in the ‘Super Saturday’ post is also intriging. Thanks Poss.

  2. Rocket said

    Nelson must be wishing the Reserve bank could lift his “Preferred PM” rate to 7.25% !

  3. nobby said

    stop the presses,nelson claims underdog status “gasp!”
    turnbull says “he needs a fair whip on the crack”

  4. Aspirational Aspirationalist said

    I can’t see the Nightwatchman leaving the field until atleast July when the new Senate comes in. Then we they might have a new straw poll and see who the leader will be. Hopefully the amount of brown sandwitch eating has decreased.

    If i was in charge at the Liberal Party HQ at the moment, i’d have a long list and be picking my superstars to place in any upcoming by-election (bye-election???) and quietly be doing a bit of research in case of a double dissolution. I’d aim so that any of them would be able to take on the pointman role and not break a sweat.

  5. Thomarse said

    bi-election?

  6. Jason said

    I like the Nightwatchman thing, Poss, but correct me if I’m wrong, but you’d usually try to stick someone in who averages in double figures – otherwise they might not make it to stumps, and you might wind up throwing another wicket away.

  7. gandhi said

    …but faced with Rudd taking the new ball late in the day, rather than the Libs looking for a middle order rally with their Turnbulls and Bishops – they played it defensive and sent in the nightwatchman; Brendan Nelson strode to the wicket.

    C’mon possum. If we are going to be totally honest with this forced analogy, the Match was already over. In fact, the Series was already lost. The Libs were playing for pride.

    The Selectors were in a quandry. Half the team was either injured or banned from further play.

    The remaining players were squabbling openly over media-apportioned blame for the long season’s debacle. All eyes were on Test spots for the coming tour abroad (far from the maddening crowd). Step forward Brendan Nelson, proudly declaring that he is playing for the pride of his insular nation, shaming others into silence in the process.

    First ball is a Golden Duck. But the umpire, perhaps stung with pity, declares a No Ball.

    Second ball is a defensive block. Small sections of the crowd applaud half-heartedly.

    Third ball is a single! Sections of the crowd leap to their feet and cheer wildly. Others are already heading for the exits…

    As the evening shadows move across the pitch, the Nightwatchman remains on 3 runs from 187 balls. Small sections of the crowd applaud every defensive block as if it were a six. Most other seats are empty.

    The Captain of the opposing team appeals for bad light, but the media-appointed umpires will have none of it – there are still dozens of overs to be bowled before Sky Sports moves to the World Badminton Finals.

    Meanwhile, on the far side of the world, a million people are dead.

  8. Aspirational Aspirationalist said

    Continuing on from Ghandi at #7

    And now one of the players is saying the people that pay and come to the match to barrack for their team should decide who the captain is.

  9. Aspirational Aspirationalist said

    sorry. Gandhi at #7

  10. EconoMan said

    Possum, I don’t want to spoil the fun about “statistically significant”, and I certainly don’t want to give Shamahan any succour, but:

    The 3% margin of error Newspoll claims, which translates as you say to ~1.6% at single digits, is based on pre-weighted data. I.e. just sample error on 1140 respondents. However, Newspoll use population weights. Therefore, because these weights have their own ‘error’, the true post-weighted margin of error is likely around 40-50% higher than the reported error.

    http://jackman.stanford.edu/blog/?p=414

    PS. It seems it’s been decided, but I must say I prefer “The Locum” to the Nightwatchman.

  11. Possum Comitatus said

    Eco, one of two things is occurring here; either

    a) Newspoll is telling porkies about the true size of their maximum margin of error. If that is the case, then the current true error margin on Nelsons ratings would be larger than the approx 1.5% given here, but so too would be the maximum error margin given by Newspoll itself. At a stated maximum of 3%, that margin is virtually indistinguishable from the size of the sampling error that would occur through pure unweighted random sampling with that sized population. If Nelsons 1.5% approximate margin is undercooked, so too is the maximum 3% margin stated by Newspoll.

    b) Newspoll weights are very small, which means the 3% maximum margin given by Newspoll is approximately true, which makes the margin on Nelsons current rating of about 1.5% approximately true as well.

    As we cant see the raw data, we can only guess – but either way Nelsons MoE is smaller than the broader Newspoll maximum MoE.

  12. EconoMan said

    Possum, taking pop weights into account, yes the maximum MoE should be bigger too — around 4%. I’m not disputing the statistical fact that the MoE is smaller at the ends than the fat 50/50 middle.

    So either of two things is occurring here:

    a) Shamahan is a hack that doesn’t understand statistics, so at a true margin of error of 3% at 50/50, he was wrong about the MoE on The Locum’s PPM.

    b) Shamahan is a hack that doesn’t understand statistics, because he claims/parrots that the Newspoll margin is 3% based on simple random sampling with n=1140, but Newspoll doesn’t simply random sample (they also weight the data for age, gender, region, and probably education, income, others?).

    (Note that if they don’t weight the data, thereby inflating the true MoE, then the poll is crap anyway. It tells us what proportion of the group sampled are going to vote which way, but it doesn’t translate to a nation (or region) wide indication.)

    ‘Newspoll is telling porkies…’
    I think it’s fair to say they’re not telling the whole truth, but then again, how many people would understand it? From Jackman (link above):

    So, when Newspoll says 58% 2PP for Labor with a “maximum margin of error” of 2.3 percentage points, what they should say is “under simple random sampling (which we didn’t do) of 1,700 respondents, a 95% confidence interval around our estimate of 58% ranges from 58-2.3 to 58+2.3 = [55.7, 60.3].” Given the argument above, that should really be something more like [55.2, 60.7], applying a design effect of 1.4.

  13. Thanks a lot for sharing this with all of us you actually realize
    what you’re talking approximately! Bookmarked. Kindly also talk over with my website =). We could have a hyperlink alternate contract among us

  14. I love your blog.. very nice colors & theme. Did you make
    this website yourself or did you hire someone to do it for you?
    Plz respond as I’m looking to construct my own blog and would like to know where u got this from. many thanks

  15. Howdy! I simply would like to give a huge thumbs up for the nice info you will
    have right here on this post. I can be coming again to your blog for extra soon.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
Follow

Get every new post delivered to your Inbox.

Join 57 other followers

%d bloggers like this: