Tuesday, December 31, 2019

2019 Site Review

This post presents site data for 2019.  The activity graph tells the story of the year (the units are unique pageviews per year):



Hmm, also looks a fair bit like my income pattern for the year.  There were three big events within two months early in the year - the NSW state election, the Tasmanian Legislative Council elections and the 2019 federal election, but the back half of the year has been extremely quiet, bar a small flutter of interest in September that was partly caused by an unusually exciting Tasmanian recount.  It is not just that the second half of the year was such a quiet one, but also the 2019 federal election aftermath was considerably less interesting than either 2013 or 2016, with only one remotely close Senate race and no House of Reps recounts (among other things).  As a result of all this, site traffic was down 22.5% on 2018 (which was the busiest year here to date) and also down 5.1% on the previous federal election year, 2016, making this year the third best for site traffic so far.  


I greatly appreciate the help from people who donate to this site, which this year covered my time for a number of weeks around the NSW and federal elections.

In 2019 I released 79 articles including this one.  The near-total disappearance of voting intention polling following the 2019 polling fail dragged my output down compared to what it otherwise would have been.  Major article subjects were the federal election excluding Poll Roundups (20), federal poll roundups (9) and the NSW state election (8).  

The cutting-room floor

Pieces started this year but that did not get finished have so far included:

* a piece on the 2019 Tasmanian bushfires, designed to correct various nonsense about the areas of certain kind of habitats burned but that didn't get very far before other things distracted me.

* an article to be called I Think We Should Talk About Section 40, which would have argued that the Speaker should be able to vote to create ties not just to break them (to avoid irregularities like the passage of Medevac laws, that would have been tied and therefore lost has the Speaker been allowed to vote with the Government against them.)

* a piece on whether the Nationals should run a separate Senate ticket in NSW.  This got nearly finished; the answer was to be that it wouldn't help since the Nationals' share of the Coalition vote in NSW was not high enough to activate the scenario in which it did.  

* a further response to the 2019 polling failure that was to be entitled I Aint Predicting Anymore.  The flavour of this response will become clear enough in my polling-based commentary on other coming elections.

* speculation, as of July, about whether an early Tasmanian state election was or wasn't plausible.  (The government was experiencing instability because of Sue Hickey, but soon had its fortunes restored when Madeleine Ogilvie was elected on a recount.)

* comments about the size-of-Tasmanian-parliament debate.

* a piece correcting misapprehensions about the previous federal parliament's resolutions about Section 44 referrals.  (These did not, contrary to some hotheaded media reporting, prevent an MP being referred to the High Court if a disqualifying factor was discovered after the election.)

* a piece to be entitled "What sort of government would I enjoy?", which was inspired (before I decided what goes on on Twitter usually stays there and lost interest) by someone on Twitter sarcastically saying "enjoy the government" during the banter after I was finally blocked by Van Badham.

There were also some others that appeared in reshaped form as parts of later articles or that I have not listed above because I think I will use the ideas in them in future.

Despite having too much free time in the second half of the year, I've had some difficulty with the motivation required to get some articles over the line, but as can be seen, plenty of them have still made it.

Top of the pops

As measured by unique pageviews, these were the ten most visited articles in 2019:

1. 2019 New South Wales Postcount: Legislative Council

A (distant) second in the history of the site on unique pageviews, and a close-ish second on pageviews overall, this article followed the extremely complex NSW Legislative Council postcount, which proved to be a fascinating and very challenging exercise in trying to project from information that was in many ways incomplete, skewed or flawed.  

This count involved an eight-party race for four seats, in which five of the parties fighting for seats were having their above the line votes included in a rough count and three were not.  Moreover two of those included were being overcounted as a result of counting errors - a point that only became clear rather late in the piece and would have some major implications.  Throw in major changes in preferencing driven by the Senate count, and variations in minor party votes between different areas and types of votes, and what you have is one of the messiest postcounts Australian psephology has ever tried (and not always succeeded, but heck, we were much better than the ABC website) to make sense of.  In the end, the Coalition, Labor, One Nation and Animal Justice were successful and Liberal Democrats (from a seemingly strong position), Christian Democrats, Keep Sydney Open and Sustainable Australia all failed.  Thanks again to Ross Leedham for his excellent work on this one.

2. How To Make Best Use Of Your 2019 Senate Vote

Voting advice resource page, slightly updated from 2016 version.

3. The Miracle Is Over: The 2019 Australian Federal Election Poll Fail

Every national Australian opinion poll failed to predict the correct winner at the 2019 federal election, and results in the last few weeks were bizarrely clustered in a manner suggestive of either herding or a combination of "self-herding" and coincidence.  Incidentally on the day after the 2019 election I had a record number of media calls for one day: eleven (my previous record was seven).  Nearly all concerned this issue, which saw me get media calls from as far afield as Japan and Argentina.

4. Legislative Council 2019: Nelson, Pembroke and Montgomery Live And Post-Count

Postcount thread for the three Tasmanian Legislative Council seats.  By total pageviews this is the third most visited page in the site's history, doubtless because of repeat viewers hitting refresh.  The Pembroke and Montgomery counts were easy retains for the Labor and Liberal incumbents but Nelson turned into an epic cutup with left-wing independent Meg Webb snaring the traditionally conservative seat from third place on primaries, only the fourth win of a Council seat from third in the last 100 years.

5. Tasmania Senate 2019: Prospects and Guide

Resource and prospects page for the 2019 Tasmanian Senate contest.  The page correctly predicted two Labor, two Liberal and one Green wins but was rather equivocal about Jacqui Lambie (who in the end won easily.)

6. 2019 House of Reps Postcount

Main page for House of Representatives post-counting, especially the close counts in Macquarie and Bass.  Macquarie looked headed for a recount before Susan Templeman (ALP) pulled a rabbit out of the hat on dec prepolls and absents.

7. Legislative Council 2019: Nelson

Preview page for the above-mentioned Nelson contest, a very strongly-contested vacancy left by the retirement of 24-year incumbent Jim Wilkinson.  The most popular Legislative Council guide page here so far.

8. 2019 Senate Postcount: Main Thread

Followed a generally quiet Senate postcount, with special attention on modelling the BTL votes for Lisa Singh and Jim Molan.

9. 2019 Queensland Senate: Who Will Be Last When The Music Stops?

The only competitive Senate postcount involved the LNP, Labor, Greens and One Nation fighting over three Senate seats.  From a beginning in which Labor seemed like they might have some hope on the more generous assumptions available, everything went against them both in late counting and on preferences and in the end Labor's Chris Ketter was the one who couldn't sit down in time, by a large margin.

10. What Are The Prospects For A Labor-Green Senate Majority?

Written at a time when Labor looked like modestly winning the election, this article thought Labor and the Greens together would in that case probably pull up about two or three seats short of a combined Senate majority.  In the end they pulled up four short with Labor narrowly losing, so given the granularity of Senate contests, the conditional prediction was probably right.

Some other stats

The ten biggest days of the year were: May 19, May 18, May 17, May 4*, May 20, May 21, May 16, May 14*, May 22, May 5*.  All were purely federal election related days except for those marked *, on which the Tasmanian Legislative Council was part or most of the action.  March 24 (day after NSW State Election day) missed the top ten by three unique pageviews.

The most popular pieces written in any previous year were How Often Are Federal Newspolls Released?, Why Preferred Prime Minister/Premier Scores Are Rubbish, "Margin of Error" Polling Myths, the increasingly ancient bio page, and Why Is Seat Polling So Inaccurate?  

The ten most clicked tags were Legislative Council, Tasmania, 2019 Senate, Nelson, pseph tied with Senate tied with silly greens, Ehrlich Awards, 2019 federal and debunkings.

The top ten visiting countries this year were the same as last year, apart from the order of the bottom five being shuffled: Australia, USA, UK, NZ, Canada, Japan (+2), Germany (+3), Singapore (+1), India (-2), and France (-4).  125 "Google countries" visited in 2019 and 168 have now visited in total.  The latter figure is again lower than 2018 and it's possible there is some issue with long-term data loss or revision in Analytics (since Gabon and Cook Islands, listed as visitors in previous years, now supposedly have visits only from 2019).  Apparently genuine first time visits were recorded from Bhutan, Madagascar and Malawi.  Madagascar had been the second most populous country to never arrive, and the four most populous non-visitors are now Democratic Republic of Congo, North Korea, Niger and Burkina Faso.

The top ten cities this year were Melbourne, Sydney, Hobart, Brisbane (+1), Canberra (+2), Perth (+2), Adelaide (-1), Launceston (-4), London (+1) and Gold Coast (re-entry).

Among the more unusual queries to reach this site were:

dr satyapauls case april 8

how many votes does fraser anning need?

how often newspoll votes done

is it best to preference someone who has no hope in a federal election

is there legislation about voting in the council election in hobart

should we use hare clark in queensland

too much information impeding performance

what do you call the leader of the Hobart consul

what does 'adrian bold' aka 'nathan carswell' of MWCC look like

What is your point of view before going to keep your vote to a particular party

why can't i get a 2019 senate ballot order of party placement in order to determine my voting preferences?

is a fox widespread or isolated

The prize for the best misspelling of my name to make it here was a tie between "kwvib bonham" and "kevin bonthron".

The top ten hit sources were Google, Twitter, Facebook, pollbludger, Bing (+1), Tally Room (-1), duckduckgo (+2), Reddit (+1), The Conversation (re-entry), Chesschat.  Ignoring the three search engines the next three were Mark The Ballot, The Guardian and the Something Awful forums.

Orders of the year

Action this year will include the Tasmanian Legislative Council elections for Huon (conservative independent recontesting) and Rosevears (left-wing indie retiring) in May, the NT election in August, and the ACT and Queensland elections in October.  It looks like a fairly quiet year (especially the first half) but some other stuff is bound to turn up.  

In the first few weeks, at some stage I predict I will post the 2019 Ehrlich Awards for Wrong Predictions - with some trepidation after a year in which poll-based psephology in general flopped. 

As a special new feature I will also award the inaugural Grand Gerry for the person or group who I judge to have made the most obnoxious contribution to electoral reform debates in 2019.  (The Grand Gerry is confined to Australia unless there are no quality Australian entries in that year and can be awarded on the basis of failure to pass necessary reform, supporting bad reform proposals, or simply presenting bad arguments for proposed reforms.)

2 comments:

  1. 'I Aint Predicting Anymore.'........nooooo say it ain't so, Bonython! One massive fail for psephology doesn't invalidate your inestimable contributions!

    ReplyDelete
    Replies
    1. I'm not going to stop doing most of what I do, and the poll failure won't affect how I cover counting and post-counting. But in a lot of cases there will be more statements of the form "if the polls are somewhere near accurate then the likely range of results is (blah)" and fewer of the form "based on the past history of polling, (blah) has such-and-such chance of happening."

      Delete