The School of Athens

The School of Athens
The School of Athens by Raphael (click on picture to view short documentary from Columbia University)

Friday 9 September 2011

It's time for a cup of tea, a Bex and a good lie down


Colleagues and scholars from coast to coast, across Bass Strait and all the ships at sea.

Dateline: Australia, Federal Politics 2011.

Once again the political world's tendency to make definitive statements about future outcomes has been on display.

Yet again, presumably due to last week's events and the release of this week's Newspoll, we have been assured that we are seeing the certain demise of the Australian Labor Party, which will be obliterated at the next federal election and will be out of power (at both a state and federal level) for a generation at least.

This pronouncement is coming from the same political world that told us the same things about the Liberal-National Coalition following the Federal Election in November 2007, when the ALP held power federally and in every state and territory; the same political world which looked at Newspoll in October 2009 (just about two years ago) and saw the ALP leading the Coalition 59 per cent to 41 per cent two party preferred and was convinced we would see a Coalition wipeout at the 2010 Federal Election; and the same political world which has been making the same definitive and usually incorrect statements for the last forty years, which I wrote about on 7 April this year, here.

It really is time for the political world to have a cup of tea, a Bex and a good lie down.

While in the last two weeks the political world became very excited (as it generally does) making all sorts of predictions, the best real-time objective measure from dispassionate observers that we have of the probability of the outcome of a future event is the betting markets.  

It suggests in the last two weeks the probability of a Coalition victory at the next federal election increased from 71 per cent to 75 per cent.

That's it.  A whole 4 per cent.  There was a whole lot of hootin' and a hollerin' for a 4 per cent change.

That means if the election were held four times the Coalition would win three times and the ALP would win once (though it is abundantly clear there are many who would be astonished that the ALP is given any chance at all "given the state of the polls").

Why should we pay attention to the betting markets?

I wrote the following in the Australian Financial Review at the start of the 2010 Federal Election campaign (when the probabilities were the reverse of today and it was the ALP who had a 75 per cent chance of winning and the Coalition a 25 per cent chance):

"As opposed to an individual view offered with little conviction other than an opinion, the betting markets are a reflection of a wide range of opinions that absorb all available information, public and private.
"It is a good reflection of the potential election outcome as the prices, and thus the probabilities, are determined by the weight of money and the forces of supply and demand, which results in an equilibrium of the relative probability of each party winning and therefore best approximates the true chance."
"However, it is important to recognise that a 75 per cent chance means the ALP has a 75 per cent chance of winning; it does not mean it will win in a landslide and win 75 per cent of the seats, nor does it mean the ALP cannot lose."
"The Coalition's chance of winning is 25 per cent, which suggests it has a real chance - just one-third as large as the ALP's."
The same applies today, just in reverse.

What has happened in the political world is that people have made judgements with absolute certainty, based on some pieces of information available today, about how 14 million people will behave in 2 years time.

Whereas the betting markets are telling us that in reality the potential political outcome is not certain and can be explained by this simple example.

Assume we have a bag of four balls - three blue balls and one red ball.  The blue balls represent the Coalition and the red ball represents the ALP.

If you randomly select one ball from the bag, you have a 75 per cent chance of selecting a blue ball and a 25 per cent chance of selecting a red ball.

They are your current election probabilities.  It is entirely possible that the ball you pull out will be a red one, in other words an ALP victory.

However, the political world is saying, "as far as we are concerned there is no red ball in that bag, just four blue balls."

Very well, if we were to assume that the view from the political world is correct and that the Coalition does not have a 75 per cent chance but rather a 100 per cent chance (or near enough), then there is a very simple test to determine whether this view is held with any real conviction.

If one of the major four banks were to offer a term deposit paying a 28 per cent interest rate for a two-year term, it would be hammered in the rush of people trying to invest.  Millions of people would be putting all their available cash, thousands of their hard earned dollars, into that term deposit because the rate is so high and the chance of being repaid in full is certain.

Now, you can still get a price of $1.28 on the Coalition to win the next election.  That is a 28 per cent return for a two-year investment (a much shorter period if you believe some of the commentary).

Would you be as confident of placing thousands of your hard earned dollars on the Coalition to win as you would by investing with the bank?

If the answer is no, then there is something that is holding you back.  And that something is the fact the Coalition, despite all the media hysteria, is still only a 75 per cent chance of winning, not a 100 per cent chance.

The next federal election is most likely to be held in October/November of 2013.  There are three years of football finals to be played before that election is held.

Who is confident of predicting which teams will win the AFL and NRL football competitions in 2011, let alone in 2013?

There is a widespread view that says "I cannot see how the ALP can win the election" and that of course makes sense, because if it could be seen then the ALP's chance of winning would not be 25 per cent but maybe 50 per cent or even 75 per cent.  However, the fact that it cannot be seen does not mean that it cannot happen.

Similarly, to suggest the Coalition cannot lose is like saying the champion racehorse Black Caviar (featured on this week's Australian Story on the ABC) cannot lose.  

But even Black Caviar is not priced at a 100 per cent chance.  

At her last start her starting price was $1.14 which is about an 80 per cent chance of winning (excluding the bookmakers' betting margin), which indicates if the race was held 10 times, she would win 8 times and lose twice.

In my last post I wrote about the champion racehorse of the 1930's, Ajax, who had won 18 consecutive races (Black Caviar to date has won 13 consecutive races) and started in the 1939 Rawson Stakes at Rosehill Racecourse, in a three horse race, at odds of 40/1 ON (a 98 per cent chance of winning) - and lost see here.

Probabilities are just that, probabilities.  They are estimations of the chance of a likely outcome.

Currently, the most likely outcome of the next federal election is that the Coalition will win.  Its chances of winning have increased from 71 per cent to 75 per cent in the last two weeks.  

A 4 per cent increase in an estimated winning chance is a rational, measured and sober reaction to the last two week's political events.

The political world's hysteria was an irrational, feverish and intoxicated reaction to the last two week's events.

It really is time for the political world to have a nice cup of tea, a Bex and a good lie down.





UPDATE 1 September 12: Samantha Stosur was priced at $5.00 to win the US Open tennis title, and Serena Williams was priced at $1.17. That meant Stosur was given a 19 per cent chance of winning (excluding the bookmakers' margin) or about a 1 in 5 chance - and won.  

UPDATE 2 September 18: The Ireland Rugby Team was priced at $5.60 to win the match against the Australian Rugby Team, which was priced at $1.17. That meant Ireland was given a 17 per cent chance of winning (excluding the bookmakers' margin) or about a 1 in 6 chance - and won

UPDATE 3 October 2: The Tongan Rugby Team was priced at $15.00 to win the match against the French Rugby Team, which was priced at $1.02.  That meant Tonga was given a 6 per cent chance of winning (excluding the bookmakers' margin) or about 1 chance in 16 - and won.

A 1 in 5 chance (as Stosur had) or a 1 in 6 chance (as the Ireland Rugby Team had) or even a 1 in 16 chance (as the Tongan Rugby Team had) - are real chances, they aren't simply numbers on a page.  

The 1 in 4 chance currently given to the ALP to win the next election is just as real a chance.


1 comment:

  1. Lulz A cup of tea and a good lie down sounds like excellent idea....But I'd be bloody amazed if you could even GET a Bex these days. (Kidney failure and all that jazz)

    ReplyDelete

Blog Archive

Our home

Our home
Earthrise over the moon (click on picture to view film)

The pale blue dot

The pale blue dot
Earth viewed from Saturn (click on picture to view film clip)

Our neighbourhood

Our neighbourhood
The Solar System (click on picture to view film)

Our Home Galaxy

Our Home Galaxy
The Milky Way (click on picture to view film)

A sister galaxy

A sister galaxy
Andromeda (click on picture to view film)

Another sister galaxy

Another sister galaxy
Triangulum (click on picture to view short film clip)

The Local Group of Galaxies

The Local Group of Galaxies
Our Galactic Neighbourhood (click on picture to view film clip).

Our farthest view of the Universe

Our farthest view of the Universe
Hubble's farthest view (click on picture to view film clip)

The virgo super cluster of galaxies

The virgo super cluster of galaxies
Galaxies within 100 million light years (click on picture to view film clip)

Galaxies within 1 billion light years

Galaxies within 1 billion light years

Universe

Universe