Oz Politics Blog

On Boxing Day, when Centrebet had Labor paying $2.20 for a win, an opportunity for arbitrage existed. If (on that day) you had placed $456.79 on Labor at Centrebet paying $2.20; and you placed $543.21 on the Coalition at SportsBet paying $1.85; you would have won $1004.94 regardless of which party had won the 2007 Federal election.

For an outlay of $1000 you would have recovered a guaranteed $1004.94. Okay, it would have only been a 0.49 per cent return on investment and you would have to wait until late 2007 to collect. And, yes, you would have done better putting your money into a savings account.

Nonetheless, arbitrage suggests a market inefficiency, where at least one of the prices is out of whack. Again, I am intrigued by the irrational betting that has occurred on Centrebet over the past fortnight. Was it incompetent betting or a feeble attempt to manage the perceptions of those like me who watch the betting market?

Yesterday I reported on the odds at Centrebet and IASBet. Subsequently, I found this statement from Gerard at Centrebet (dated 22 December),

Just when we thought that the Coalition had plenty of in-house turmoil that might affect them at the next election up bobs a punter who fancies them to get back in.

We eased the Coalition out to 1.85 equal favourites with Labor on the back of John Howard’s workplace reforms, and while those reforms have been receiving plenty of criticism, one of our election clients disagrees.

This week he placed $10,000 at 1.85, $20,000 at 1.75, then another $20,000 at 1.65. That’s where the price remains for the moment and it does appear very short odds given what has been going on. Labor meantime has drifted out to 2.10.

Of note, other bookmarkers are not reflecting the movement seen at Centrebet and (to a lesser extent) IASBet.

• SportsBet still has it as even money, paying $1.85 for both a Coalition and Labor win.
• SportingBet is paying $1.80 for a Coalition win and $1.90 for a Labor win.

It was enough to make me wonder whether the movement at Centrebet was a ham-fisted attempt to influence the odds (and consequently public opinion) by (for example) a Coalition support group. Certainly, it was not rational behaviour. All other things being equal, the self-interested punter would not have placed the second and subsequent bets at Centrebet when other bookies have better odds on offer. Strange!

If Coalition supporters are gaming the betting market in order to change public opinion about Howard’s chances next election, it could prove expensive as we approach the next election.

Changing topic, also of note, SportingBet and SportsBet have books on who will lead the Liberal Party at the time of the next Federal election. This morning’s odds were:

It would not have been a Merry Christmas in the Beazley household. Both Centrebet and IAS have firmed for the Coalition.

• For a Coalition win Centrebet would pay $1.60 and for a Labor win it would pay $2.20. The implied probability of a Coalition win at the next election is 58 per cent. Update at 9:30pm on 27 December the odds have come in a little: $1.65 and $2.10.
• For a Coalition win IASBet would pay $1.75 and for a Labor win it would pay $2.05. The implied probability of a Coalition win at the next election is 54 per cent.

Morgan has released the last poll results for 2005. On the weekend of 17-18 December 2005 they polled 887 voters. Its headline predication was a national two-party preferred vote of 52 per cent for Labor and 48 per cent for the Coalition. Such a national vote would translate to a comfortable Labor win.

The primary vote predictions were as follows:

• The Coalition — 39.5 per cent (up 1 on the previous Morgan poll)
• Labor — 39.5 per cent (down 2)
• Australian Democrats — 1.5 per cent (down 0.5)
• The Greens — 10.5 per cent (up 0.5)
• Family First — 3 per cent (up 0.5)
• One Nation — 1.5 (up 0.5)
• Independent/Other — 4.5 per cent (up 0.5)

The pollster’s interpretation:

“Support for the Labor Party, continues to drop now that the Senate has passed the Government’s electorally unpalatable Industrial Relations and Voluntary Student Unionism legislation.

“In a period dominated by the race riots in Sydney, support for the Coalition rose slightly to 39.5%. However this is still well below the 2004 Federal election result. If an election were held today the ALP would win and Kim Beazley would be the Australian Prime Minister.”

No arguments from me this time, although I suspect the Voluntary Student Unionism legislation had little impact either way on many of those polled.

We are also starting to see the bounce in the most recent polls show up in the aggregated monthly polling.

The usual graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.

The big question: If 2005 could be characterised as a new hope for Labor, will we see the empire strike back in 2006? Will 2007 see the return of Jedi Master Beazley?

Today, Centrebet changed its odds from even money to $1.65 for a Coalition win and $2.10 for a Labor win at the next Federal election. This translates to a 56 per cent predicted probability of a Coalition government after the 2007 election, and a 44 per cent probability of a Labor Government.

Also of note, the book on the Leader of the Liberal Party come the next election is not up at the moment. (Only this morning I updated an earlier post on this book).