August 31, 2004


Investors Bet for Bush Over Kerry (Reuters, 8/30/04)

Opinion polls show support for President Bush and his Democratic rival as almost dead even, but as the Republican convention began on Monday investors trading presidential futures did not believe the race would be nearly so close.

Trading in the Iowa Electronic Markets showed Bush pulling ahead of his rival, Democratic Sen. John Kerry, with a 54.8 percent probability of victory, compared to a 45.4 percent for Kerry.

The Iowa presidential futures contracts, which were launched by professors at the University of Iowa to study the forecasting power of markets, have had an average 1.37 percent margin of error in predicting the winner of the popular vote, a better record than most opinion polls. [...]

The Kerry contracts had a similar rally before the Democratic convention in Boston in July. At that time, the Bush contract showed a 50 percent probability of victory to Kerry's 49 percent. But since then, Bush has regained the lead.

Traders have placed similar bets at another online exchange, Intrade, which shows Bush at just under 58 percent to Kerry's roughly 43 percent. The Bush contract traded as low as 49 percent this summer.

Rallying to a point where you're still losing seems dissimilar to rallying to the point where you're near a landslide, but why quibble.

Posted by Orrin Judd at August 31, 2004 6:03 PM

Both markets have shown Bush as the favorite for most of the campaign. It's the difference between opining on who 'should' win and betting (real money) on who is most likely to win.

It also reminds me of my favorite exchange from the article in the NY Times about Ray Fair of Yale, and his election model (also predicting a Bush win). Here's the exchange:

Interviewer: But the country hasn't been this polarized since the 60's, and voters seem genuinely engaged by social issues like gay marriage and the overall question of a more just society.

Ray Fair: We throw all those into what we call the error term. In the past, all that stuff that you think should count averages about 2.5 percent, and that is pretty small.

Posted by: Dave Sheridan at September 1, 2004 8:54 AM