Aggregated opinion polls: November 2004 to April 2005
As foreshadowed, I have now aggregated the Newspoll, Roy Morgan, and ACNielsen opinion polls for each month since November 2004. The objective was to remove some of the volatility from the individual polls so that one could better identify trends. If the next graph is anything to go by, it appears to have worked.
While not all of the April 2005 polls are in, I have included the 4 polls so far this month as they tell an interesting story. Since January 2005, the Coalition has been on an accelerated decline in terms of its two-party preferred vote, and in April 2005 it entered negative territory. Collectively, the polls suggest that the Australian Labor Party would win an election at the moment, with around 51.5 per cent of the two-party preferred vote.
Depending on your political persuasion, this does not mean it is time to either break out a bottle of bubbly or find an unlocked window in a tall building. Opinion polls are notoriously poor at predicting future election results two and a half years out from the next election (ie. they only get it right about half the time).
However, this accelerated decline may explain why the odds at Centrebet have softened slightly for the Government since 1 April. The Coalition is now on $1.55 to Labor’s $2.30, giving implied win probabilities of 60 to 40 per cent.