April 11, 2019

Horse 2530 - I Predicted The Election Date By Looking Into My Election Prophecy Steel Can Filled With Mud

I would like to refer the reader to Horse 2395 and the prediction that I made a year and a day ago:

My inkling is that there will most likely be a May 18 2019 election, with the Appropriation Bill No.1 having been presented to the House in April.
- Horse 2395, 10th April 2018

As I bash into my tablet like a cat with a shiny thing who has somehow acquired the gift of language, the Prime Minister Scott Morrison is in C-1 and making the drive from The Lodge to the Governor-General's place to make the advice that parliament should be dissolved and that a general election should be held on Saturday May 18.
I am not just any cat with a tablet but a cat who has just been appointed Professor of Gravity Studies at Whiskers University. Allow me to knock everything off of the table in the most elegant and smug way possible. I make no bones about the fact that this post is one of gloating, self congratulation and superbia. In the words of the great Benjamin Disraeli:

I told you so¹.

The timetable for the next Senate election is determined by the Constitution. Senators are appointed on a six year term; with half of the Senate being elected every three years. As the opening date of the next session of the Senate is July 1, then the election for half of the Senate needs to be done so that everyone can assume their seats on opening day. The Australian Electoral Commission has advised for more than half a century that they would like to have six weeks to count ballot papers and give themselves time to allow a recount should there be a dispute with the results. That means to say that the absolute last day that the AEC could ask that the election be held is if May 20 were to fall on a Saturday. There has been a Senate election on May 31 but that was under an older set of voting rules where you had first past the post for multiple seats and it was easier to count.

Running exactly counter to the AEC's wishes is the desire of the Prime Minister of the day. A Prime Minister who is in charge of a popular Government would like to go to the polls earlier in the election cycle so that way they retain power. Under normal circumstances, half way popular Governments like to go to the polls in late October, November or early December because they know that the electorate would prefer not to have their footy watching times disrupted, neither do they want to hold an election during the stress of Christmas. If a Prime Minister were to declare an election for either the Australian Rules Football Grand Final weekend, or the Rugby League Grand Final weekend, or to a lesser extent the weekend of the Bathurst 1000, then the people of Australia would open up a can of SPC Baked Beans, Scorpions and Acid, and the Prime Minister would find themselves eating a dangerous meal all alone and out in the street. Not even Mr Morrison is daft enough to hold and election in September; which is a little like writing a suicide note because by itself it is worth half a percent of votes; which in marginal seats is sometimes more than enough to flip them.
The problem for the Rudd-Gillard-Rudd Government was the same in principle as for the then Abbott-Turnbull Government. We have witnessed the most effective Opposition Leader in Australian political history, in Tony Abbott. Once his master (John Howard) had been unseated from his own office, this attack dog has been roaming around inside the Canberra Bubble unleashed. Tony Abbott has been such an effective Opposition Leader that he not only brought down Rudd, Gillard and Rudd again but he also brought down himself and Turnbull and has been running a pretty solid Opposition against Morrison. I don't know how long Mr Morrison could hold his government together or whether or not it would suffer another change in leadership before an October or November election. In all honesty, Mr Morrison has had his hand involuntarily tilted in declaring this date.

I work in an accountancy firm which apart from doing people's tax returns, also does a fair amount of forensic accounting. When you have a business which needs to be valued (which usually happens when partners want to escape the business, or when people in a Family Law matter want to escape each other²), then in order to work out what the business will be worth in future, you need to do a many variable trend analysis.
Watching politics is like watching sport for so very many reasons but the relevant one here is that there are ready statistics on polling data and sentiment, which you can then run the same kinds of trend analysis for. Perhaps the most famous case of political trend analysis going wrong according to the world was for FiveThirtyEight which predicted a Hilary Clinton win for the Presidency of the USA. What people fail to realise though is that it always had the likelihood of Donald Trump winning at about 30% - 35% and it kind of got the raw vote data about right. I merely ran trend analysis on the data from Ipsos, Galaxy and Newspoll and extrapolated it out to beyond the end of the last possible date of the House of Representatives and the Senate; and came up with doom for the Government.

At the time, I had been looking at at least 30 sets of polls and concluded that the can needed to be kicked down the road as far as possible. Today is just the other end of that road. What I didn't forsee and couldn't have forseen is that the Abbott-Turnbull-Morrison Government has been so much of an omnishable and completely unable to govern itself much less the country, that 30 unfavourable polls would blow out to more than 50. There has been no bounce off the bottom of the barrel, this government has just stuck itself to the crud down there and stayed put.
Predicting the last possible date for the election was sensible. Predicting the knifing of another Prime Minister was impossible. At this point, predicting a Labor Government seems like a fait accompli.

Also, before anyone thinks that I am somehow a prophetic genius, the only reason that this post exists is because I it right. Had I got it wrong, I would have said diddly-squat. That's the thing with nonsense prophecy, it's really easy to look like a genius after the fact by only selecting favourable results.

¹I actually don't know if he said this or not but he was the kind of magnificent knave to have done so.
²In at least 97% of cases it is the husband's fault. 

No comments: