Prediction Markets Fail Again

GO FIGURE: Prediction markets — those thinly traded games played by college kids and other traders  who lack the capital to trade in deeper broader markets (equities, fixed income, commodities, currencies) — are for shit:

"To the amazement of economists and online bettors, the answer has varied a great deal among betting Web sites.

Markets are not supposed to work that way, even online prediction
markets, where bettors trade on the chances of a candidate’s winning an
election in the same way that they might bet on pork bellies to go up
in value.

In the last few weeks,, which is based in Dublin, had
consistently given John McCain as much as a 10 percentage point edge in
his chances to be elected president compared with other large online
overseas betting sites. These include the British-based, as
well as the Iowa Electronic Markets, a research project at the
University of Iowa that allows bets of $500 on election results."

Actually, thinly traded markets such as this are supposed to work EXACTLY this way. What prevents the "real" markets from operating this way (most of the time) is the enormous amounts of money at stake, and the huge and diversified crowds of traders watching for aberrations. Little markets, small amounts of money (millions not trillions) and thin trading are prone to this sort of nonsense. (Don’t say you weren’t warned)

Here’s, the question of the evening: How many legitimate market strategists, economists, and traders still believe in the Efficient Market Hypothesis?  Anyone?


Misunderstanding Prediction Market Failures (February 14, 2007)

Why Prediction Markets Fail (January 11, 2008)

Trading Variance in Election Predictions Raises Questions
NYT, October 19, 2008

Intrade Betting is Suspicious
Nate Silver,
Five Thirty Eight , September 24, 2008

Trader Drove Up Price of McCain ‘Stock’ in Online Market
Josh Rogin,
CQ Oct. 17, 2008 – 3:44 p.m.

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:

Posted Under