Blockchains as Arbiters of Truth
Cryptocurrency Ethereum Finance
Subscribe to our newsletter

Join our mailing list for the latest news, exclusive offers and community events.

article avatar
Polling people on whether they think outcome A or outcome B will happen is not very effective in determining which of the two actually will happen. Polling is a poor indicator of future events for a variety of reasons: people with more expertise are given the same weight as those with little to none, people are biased, and not incentivized to give the correct answer. Even asking people about past events or facts are heavily influenced by their cultural and political leanings.

For instance if you were to ask people whether gun production increased or decreased during the Obama administration, the results are skewed by their party affiliation. (It increased 192%). However, paying people to give the right answer, or not answer at all, greatly increases the accuracy of the result.

For these reasons, prediction markets—a financial market on whether or not a particular event will occur— are more accurate than polling and other forecasting techniques, e.g. which movie will win best picture? will there be a second referendum on Brexit? will the FED raise interest rates? They’re more accurate for the simple reason that those in-the-know are incentivized to take a position, and take a more vested position, the greater their knowledge and confidence.

Many companies have used prediction markets internally for forecasting when a new product will launch or to predict sales. Simply asking managers for target dates, results in answers that are too optimistic—they’re motivated to be positive. With a monetary incentive (the more right you are, the bigger the bonus), more accurate forecasting can be made. Hewlett-Packard, Siemens, and Best Buy have all used internal prediction markets to forecast: sales, delivery date of products, store opening dates, and whether new services will be introduced on time. They outperform the companies’ traditional forecasting tools by up to 70%.

While some prediction markets do exist, they have failed to gain mass market traction. Gaining traction is hard because of the need for trust in the company making the market. There is no transparency that it’s set up fairly, that the company is not taking unfair advantage or users, nor guarantees that they will pay out the winners. In short, it’s a problem of trust.

This is where the blockchain comes in. The blockchain is the perfect platform to host the formulas and money of the markets. The code can be vetting and the payouts guaranteed to match what’s expected. Augur, Gnosis, and Stox are 3 prediction markets on Ethereum. In addition there are now user-friendly apps that are being built on these platforms. Guesser is an easy-to-use app that uses Augur smart contracts behind the scenes. I used it to make a prediction on the oscars. I predicted wrong.

For a comparison of Augur, Gnosis and Stox see here. To read more about Guesser see here.

For a good overview of prediction markets in general see this paper.