Harnessing the Power of Prediction Markets
By Brandon BeckhardtIn an era of evolving uncertainty and near-infinite data, it’s becoming increasingly difficult to sift through the noise and paint a picture of what the future holds. Prediction markets have emerged as powerful tools to help paint that picture, giving prediction market followers another tool in their arsenal to forecast the likelihood of future events.
How do Prediction Markets work?
Prediction markets operate on the principle of collective intelligence and utilize a trading system to generate forecasts. Participants trade contracts, whose price represents the likelihood of specific events occurring. On Kalshi, this is done by users trading event contracts. To learn more about event contracts, read our article What is an Event Contract?
The contracts’ value fluctuates based on what the market perceives the probability of that event occurring to be. As participants buy and sell these contracts, the market adjusts, and prices converge toward the most widely accepted probability. Market prices, therefore, serve as an aggregated prediction of the event's likelihood. The market’s assessed probability, and therefore contract price, may change over time as participants learn more or events on the ground change.
The Wisdom of Crowds
Prediction markets leverage the "wisdom of crowds" phenomenon, which suggests that collective opinions and judgments tend to be more accurate than individual assessments. By aggregating the knowledge, information, and insights of a diverse group of participants, prediction markets harness the collective intelligence to generate forecasts.
Decades of empirical evidence show it is nearly impossible to consistently beat the market (though there are a few outliers). In the case of prediction markets, the assets may differ but the basic principle remains unchanged - the market consistently outperforms most any individual.
Limitations of Prediction Markets
While Prediction Markets offer many benefits in forecasting the future, they are not a panacea. There are areas where prediction market forecasts are limited.
Black Swans are Black Swans because they are nearly impossible to predict. The fat-tailed nature of the many improbable but impactful events that can happen does not fit well in the model of markets. Taleb and Tetlock discuss this in their joint paper and cite revolutions as an example event where prediction markets can be erroneous.
It is notoriously difficult to forecast events extremely far out, as Tetlock outlines in Superforecasting. Prediction markets may do better than an individual, though their accuracy wanes as the event horizon gets further out.
While prediction markets can fall victim to the unpredictability of fat-tailed events and far out events, one counterweight benefit of prediction markets is that you can trace the market forecast over time - this quantitative measure of the market sentiment over time can help combat hindsight bias that so commonly plagues those looking at history in the rearview. As Taleb mentions in Black Swan, it can be far too easy to gloss over the inherent unpredictability of past events and engage in retrospective storytelling that overemphasizes causal relationships and makes events appear more predictable and more linear than they actually were. This human tendency to create coherent and simplified stories about past events to make them more understandable and predictable can be partially counteracted by prediction markets.
Advocates of Prediction Markets
- Nassim Taleb highlights specific domains where forecasts given by prediction markets can be valuable. Many of Taleb’s arguments, throughout his insightful Incerto and other work, center around the importance of having skin in the game when making forecasts.
The notion of skin in the game is a core tenant of prediction markets, as it is having money at risk that incentivizes participants to maximize their likelihood of being right. It also washes out participants who tend to be wrong over time. - Taleb’s coauthor on the paper above, Philip Tetlock, author of the regarded book Superforecasting highlights key elements of forecasting and the power of prediction markets. He wrote the book after conducting one of the most comprehensive studies ever conducted on human forecasting potential. Read our summary of the 10 Commandments of Superforecasting highlighted in the book.
- Larry Summers has written about the aggregate accuracy of market participants, citing studies showing “prediction markets do a better job of forecasting elections than pollsters and why Hollywood studios use such markets to judge the likely success of movies.”
- Jason Furman, the Chairman of the Council of Economic Advisers under President Obama from 2013 to 2017 and his Chief Economist extolled the virtues of prediction markets, saying that “in the White House I, along with other members of the economic team, would regularly refer to prediction markets on electoral outcomes and specific events to help inform our understanding of how political and economic developments would affect economic policymaking.” He also mentions “There is ample academic evidence to suggest that prediction markets are highly efficient at aggregating information to produce an accurate forecast when compared to alternatives."
Conclusion
Prediction markets have revolutionized the way we forecast events by harnessing the collective wisdom of crowds. Their accuracy in predicting outcomes, whether in politics, sports, or finance, has made them invaluable tools in decision-making processes. As these markets continue to evolve and grow, especially with Kalshi paving the way as the first regulated exchange to offer event contracts in the US, they hold the potential to enhance our ability to navigate uncertainty and make more informed choices in an increasingly unpredictable world.
Participate in prediction markets by trading event contracts at www.kalshi.com. Become a consumer of prediction market forecasts by checking out our inflation and fed rate market forecasts, checking our market pages, or pulling market data directly from Kalshi's API.
Read next