Saturday, September 7, 2013

Election night arrangements and election-watching hints

 Home now (midnightish): Late night wrap coming in next c. 2 hrs.


Over here folks:  http://www.themercury.com.au/news/special-features/federal-election-2013-analysis/story-fnji1hk4-1226711813183

I had hoped to post some more things here in advance of election night but unfortunately I have run out of time and need to work on some spreadsheets for this evening and also find some time to relax a bit in advance of tonight.  Therefore this is just a quick post to detail what will be going on here tonight and tomorrow.  The short answer is: perhaps not much!

Tonight I am very pleased to say I will be live-blogging at the Hobart Mercury at their website themercury.com.au.  The direct link will be posted here when known. At the moment I expect to start around 6:30 and to keep going for maybe four hours.  I will also be involved with writing an article for them tonight (or doing an interview for an article; not sure which yet) so I may not be back on here tonight at all.  If I am there will be a new late-night counting extra thread up.



I am not sure of the details of the Mercury coverage but it may be interactive.  I don't expect to be on Twitter at all this evening or until tomorrow afternoon, but may be on briefly very late at night. 

Tomorrow I will be on a field trip (as a bit of a break from the election!) in the morning but I expect to have some kind of ongoing post-count developments thread and maybe a wrap of some kind up some time tomorrow late afternoon or evening.

I respectfully ask journalists not to call me between 6:30 and 10:30 tonight as I will be extremely busy. If you read my live coverage you may find questions answered there. I may be able to talk after that. Scrutineering info from those who have my number is welcome at any time.

Some things to bear in mind when watching coverage (that is, for those who need visual input and not just my words on a screen!):

* It is likely we will know the result very early and the rough scale of the result within a couple of hours.  However we will probably not know the fate of every specific seat on the night.  Especially with a very high rate of early voting, seats that are within 51:49 or closer often change sides over the postal vote count, which takes several days.  Even seats outside that level will sometimes flip.  Last time there were no very close seats and the outcome was reasonably clear on the night, but movements of three or four seats from the on-the-night count happen sometimes.  Usually there will be half a dozen seats on serious close seat watch over coming days. Every now and then a seat goes to elaborate rechecks and takes a very long time to sort out.

* The AEC sets a notional two-party count to do on election night for each seat.  Sometimes it turns out that the real contest is between one of the parties and someone else, especially where there are Independents or minor parties challenging for seats.  There may be some cases where a seat is being given as a clear win for a candidate, but they are not winning, because the wrong candidates have been selected for comparison.  When this happens, a two-candidate count between the real contenders will occur in following days.  (If there is one of these going on, I will post some advice on watching the count in that seat.  Denison last time was a real mess because of misunderstandings of how this works.)

* The ABC's computer system frequently "calls" seats incorrectly by putting them definitely in one party's pile if one party is more than 50.5:49.5 ahead on projected vote share after a certain percentage of the vote.  Be a bit wary of its calls until seats that are remotely close have a higher vote share counted.  (In some elections with very large swings, it has even allocated seats to one side or the other in advance of counting, only for those allocations to be wrong.)

* Most pre-poll votes (those cast at prepoll centres) will be counted on the night.

* We may know useful things about the Senate later tonight, but it is likely there will be uncertainty over at least a few seats in many states.  Be especially wary of the ABC computer allocating seats to micro-parties.  Although it is likely some micro-parties will win at this election, often whether one gets up or not is very sensitive to vote share changes, and vote share changes in the Senate in post-counting have much more impact than in the House.

* We will not know the 2PP vote very accurately on the night.  We will probably know it to within a point, and projections of the final 2PP (that model off late postal voting) may be accurate to within half a point, but it will continue to change with late counting.

* A common misconception about the Senate (which even some experts get wrong) is that some people believe only above the line votes are counted on the night and that this will result in a great increase in the vote of those parties that get higher shares of below the line votes, especially the Greens.  Rather, what normally happens is that above and below the lines are counted and totalled together but are not segregated by candidate on the night.  Thus the totals you see on the night should be reasonably accurate (but the movements of a few points for the major parties can have a big impact on the outcome). 

* For any seat, voting intention varies hugely by booth and therefore a side that has a strong lead with 10% counted may not win.  What you need to be looking at is the swings in the booths (or overall).  In some cases, swings might vary between different kinds of booths.  For instance in Tasmania we may see big swings in timber town booths, which tend to report early because of their small size.  So even projections of the sort done by the ABC computer can be inaccurate when based on early counting.

* Comments provided by party folks online about election results are of variable quality.  Some of them aim to be very objective (I was very impressed by the performance of now Victorian Opposition Leader Daniel Andrews in this role at the last Victorian state election) but others will act as cheerleaders, aiming to boost morale by painting a rosier picture for their party than facts support.  The Greens are especially bad at this, though they may get competition from PUP, KAP and the like this election.

I'll probably be watching the AEC website, and the ABC coverage because lots of people will be watching it and there may be things there I need to comment on.

Hope readers enjoy my coverage at the Mercury tonight, and for those who haven't voted yet, stop reading this and get your act together! 

No comments:

Post a Comment

The comment system is unreliable. If you cannot submit comments you can email me a comment (via email link in profile) - email must be entitled: Comment for publication, followed by the name of the article you wish to comment on. Comments are accepted in full or not at all. If you submit a comment which is not accepted within a few days you can also email me and I will check if it has been received.