Here's the best discussion I could find on how much people are willing to bet of their Presidential candidate being victorious:

"The Iowa Electronic Markets, a policy market operated by the School of Business, uses $1.00 option contracts to derive probabilities of electoral (and other) outcomes. For the past few years they have been tracking primary and general election campaigns to look at predicted outcomes. Here's the final graph on today's election outcome probabilities: People are willing to spend $.90 on an Obama victory, i.e., they are assigning a 90 percent probability of an Obama win, while people are only willing to spend $.10 on a McCain victory today.

This history of this approach goes back to DARPA when then Admiral John Poindexter suggested a terrorist futures market to obtain probabilities of a likely terrorist attack. He was booed out of office for "trading on terrorism" as the Washington pundits put it back then. What is interesting to track is whether policy prediction markets gather information at a lower cost with greater accuracy than, say, traditional polls.

Five Thirty Eight (http://www.fivethirtyeight.com/) an increasingly closely followed website, uses multiple polling and regression analyses to derive composite predictions on today's elections a the presidential and Congressional levels. As the attached graph from today's update indicates, they assign a 98 percent probability of an Obama popular vote win. "

Anyone willing to put their money where their mouth is? 10:1 odds looks like a way that a ardent McCain/Palin supporter could really clean up.

Cheerily,
Charlie