Prediction Markets Improve Forecasting and Risk Management

by Erin McCune on April 9, 2008

in Caught My Eye, Management, Research Round Up, Technology


(graphic courtesy of The New York Times)

Corporate decision makers are honing their accuracy by tapping into the collective wisdom of employees and customers via web-based prediction markets. An article to today's New York Times explains:

Corporate prediction markets work like this: Employees, and potentially outsiders, make their wagers over the Internet using virtual currency, betting anonymously. They bet on what they think will actually happen, not what they hope will happen or what the boss wants. The payoff for the most accurate players is typically a modest prize, cash or an iPod.

The idea isn't new, but it is gaining traction in larger companies. The article cites efforts by InterContinental Hotels, GE, and Hewlett-Packard to improve forecasting, reduce risk, and accelerate innovation.

Starting a year ago, a group in the purchasing unit at H.P. began prediction markets on the price of computer memory chips three and six months ahead. The prediction markets, Dr. Huberman said, were up to 70 percent more accurate than the company’s traditional forecasting models. The more accurate predictions, he said, can be used to finesse purchasing, marketing and product pricing decisions.

The article includes many examples, including how Best Buy is using prediction markets to predict not only demand for specific products but whether new stores will open in time.

See also this column from last summer by James Surowiecki in the New Yorker about prediction markets used to identify best selling books and anticipate the popularity of movies. Surowiecki expands on his ideas in his book The Wisdom of Crowds (available from Amazon).

Excerpt  from Surowiecki's New Yorker column:

MediaPredict, however, is wagering that in the real world success is, at least in part, predictable, and it follows a model that, over the past decade, has proved surprisingly effective in forecasting a wide range of events: the prediction market. Prediction markets function like futures markets, except that, instead of betting on the future performance of a company or a commodity, people can bet (often with play money) on things like election outcomes, current events, and product sales. Rather than relying on the gut instincts of a single decision-maker, prediction markets tap the collective intelligence of everyone playing the market. The most successful media prediction market is the Hollywood Stock Exchange, in which traders collectively forecast the box-office performance of Hollywood films, Oscar nominations and results, and the performance of individual actors, with striking accuracy. The market on average picks more than eighty per cent of Oscar nominees correctly, and hasn’t missed more than one Oscar winner in the past four years. More important, it has also done a good job of predicting box-office performance. According to a study by Anita Elberse, a professor at Harvard Business School, the market’s forecasts are off, on average, by sixteen per cent—far from perfect, but a track record that most studio marketing departments would be proud of.

Surowiecki warns that in order to be reliably smart – smarter than the individuals in the participating group – a predictive market must have the following characteristics:

  1. Diversity of opinion to draw on different points of view, from participants with access to different information;
  2. Independence of members from one another to avoid thought leaders influencing the opinions of others;
  3. Decentralization to ensure that local and specific needs are taken into account;
  4. A reliable method for aggregating opinions to ensure that all results are taken into account, but avoids the perils of decision by committee. This is the trickiest part.

Learn more:

Betting to Improve the Odds
The New York Times
April 9, 2008

The Financial Page
The Science of Success
by James Surowiecki
The New Yorker
July 9, 2007


The Wisdom of Crowds
by James Surowiecki
Published 2004

Leave a Reply

Previous post:

Next post:

Clicky Web Analytics