With six weeks until the Queensland State Election, and it being a Presidential election year in the USA, I thought it timely to inform you that studies have shown betting markets are more accurate predictors of election results than are opinion polls or economic forecasts.
A 2005 Discussion Paper by the Australian National University’s Centre for Economic Policy Research highlighted that in the 2004 Federal Election, betting markets not only correctly predicted the Howard Government’s victory, but also the result in three-quarters of marginal seats. The 77% election-eve probability of a Coalition victory (as calculated from the odds) could also be argued to have correctly predicted the Government’s increased majority, considering the comparative percentage was 60% prior to the 2001 election.
In 2001, favourites won 43 of the 47 marginal seats on which there was betting, while in 2004 it was 24 of 33, with one seat having no favourite. “The market’s prediction of the likely winner also predicts actual vote totals reasonably well. Of those cases where the market-favoured candidate lost, the margin was generally less than a couple of percentage points”, the paper’s authors noted.
Estimates put the amount wagered on the 2010 federal election at $10 million. In the week prior to the election, the ALP firmed to $1.40, but got out to as high as $2.75 during the week of uncertainty regarding which party would form government.
In the USA, large and organised betting markets operated on every election between 1868 and 1940, according to a 2003 report by the University of North Carolina; as much as $206 million in today’s money was wagered on the 1916 poll in New York markets. The report concluded that “the market did an admirable job in forecasting elections in a period before scientific polling. In only one case did the candidate clearly favored [sic] in the betting a month before Election Day lose”.
Of the 15 US Presidential elections (held early November) between 1884 and 1940, the mid-October betting favourite won 11 (73%) and the underdog won only once (Woodrow Wilson in 1916, who had come into equal favourite by polling day). According to the report, “the betting odds possessed much better predictive power than other generally available information such as party affiliation or incumbency. Even if one takes the conservative assumption that the 1884-92 predictions were failures (when markets were relatively even), the New York betting record was more than a match to chance”.
In summary, if you back the favourite less than a month before an election, you won’t lose too often.
Click here to view the latest Election markets.
