The Iowa Electronic Markets' presidential contest gives Barack Obama a 28-point lead over John McCain, while a Gallup tracking poll released Monday has Obama up by only 3 percent.
The Democratic contract on the winner-take-all market traded at 64.3 cents, according to its Web site. This figure shows that investors believe there is a 64 percent probability that Obama will win the popular vote in the general election.
The Gallup poll had Obama up 46 to 43 percent.
Polls and, increasingly, markets have been used to judge the outcomes of elections. This year's difference means that investors seem to like Obama, while polls of likely voters show they are considering McCain more.
What is unique about the various Iowa Electronic Markets, said Forrest Nelson, a UI professor of economics who oversees the enterprise, is that they are constantly being followed.
"The polls don't get much attention until right before the election," he said.
The UI College of Business introduced the Iowa Electronic Markets in 1988 to offer a real-money online futures market.
Anyone can participate by purchasing "shares" of different possible outcomes and receive payouts if right. For example, if someone buys a share of John McCain winning the presidency - currently at $0.36 - and he is elected in November, that share pays off $1.
The theory is, with enough people predicting with their money on the line, a more-correct prediction will emerge.
With the rise of the Internet over the past 20 years, the markets have become increasingly popular.
Nelson said that, as a general trend, polls seem more erratic, whereas the markets remain stable.
Though, he said, both the polls and markets have their own purpose, and it makes sense to use them both.
"They each are within the ballpark of what could happen," he said.
The markets can be hard to understand partly because they don't know who is investing, he said. However, that if someone were to choose a single predictor of the election, markets would be better.
"Polls deal with a hypothetical situation of what would happen if the election were tomorrow," he said, noting that the presidential market predicts what would happen in November.
Ann Selzer, the president of Selzer and Co., a polling firm, said the differences between the markets and the polls is methodology. She said her firm's research involves scientific backing, whereas the markets are only based on those interested in participating.
However, she said, the polls are always facing new challenges.
"The industry has taken some hits; people like to bash them," she said.
Selzer said an obstacle pollsters face is the increased presence of online polls.
"They are not scientifically based - all of that [data] are garbage," she said.
Similar to the polls, Joyce Berg, the director of the Iowa Electronic Markets, said market officials are always working to refine the data. The markets and the polls can be viewed as scorecards.
"They just record in a different way," she said.
When the Iowa markets were introduced, Berg said, the investors were only from the UI. Today, people all around the world use the markets to predict election outcomes, she said.
"What is surprising is that as it increased in popularity, the accuracy did not change," Berg said.
The Markets' accuracy remains reputable because it involves individual people investing, she said, not a variety of pollsters.
And Berg said market officials are continually orking to find new methods to achieve more accurate predictions.
"Accuracy is always difficult to define," she said.
Iowa has a reputation for political forecasting, Berg said, noting that UI alumnus George Gallup started the Gallup Poll in 1935. So it seems natural to have a futures market for political candidates, she said.
"There must be something in the water here," Berg said.
E-mail DI reporter Anna Lothson at: