In typical stock markets, a human buyer must be matched with a human seller of stock. For example, if Joe is selling 50 shares at $50/share, there must be a buyer willing to pay that price and vice versa. The price of the stock itself is based on the supply and demand of the finite number of shares in play. In a real stock market and even most prediction markets in existence today, these transactions are all handled by computers, enabling millions of transactions like this to take place automatically each day.

In Media Predict, we do not force a buyer to be matched with a seller, and vice versa. From a trader's perspective, Media Predict is always the buyer and seller of shares and there is no limit to the number of shares in play. Media Predict also sets the stock price according to demand or lack there of. If a trader buys shares, there is demand for the stock and its price goes up. If a trader sells, there is a lack of demand and the price goes down.

The principles behind our algorithms originate from research by Professor Robin Hanson at George Mason University. Here is some background about Hanson's automated market maker:

• Hanson Paper - Combinatorial Information Market Design
• Hanson Paper - Logarithmic Market Scoring Rules for Modular Combinatorial Information Aggregation
• David Pennock of Yahoo's take on Hanson's market maker
• Chris Hibbert's view of market scoring rules