Why Prediction Markets Struggle to Rival Scientific Expertise

Polymarket's prediction of a 19% probability for a WHO-declared hantavirus pandemic dropped to 5% following initial news, illustrating how market volatility often reflects public anxiety rather than biological reality.

Sentiment vs. Scientific Modeling

Prediction markets like Polymarket and Kalshi operate on the principle of the "wisdom of crowds," where economic incentives drive prices toward true odds. However, Richard Borghesi of the University of South Florida notes that these markets are not substitutes for peer review or expert judgment, especially when traders lack specialist knowledge.

In disease outbreak forecasting, researchers typically rely on mathematical models, hospital surveillance, genomic data, and school absences. In contrast, Polymarket users often reference WHO updates and case numbers in comment sections, a behavior Harvard epidemiologist Bill Hanage suggests may be driven more by post-COVID-19 pandemic anxiety than by rigorous data analysis.

Divergence in Climate and Quantum Forecasts

Climate forecasting shows more alignment between markets and experts. A Polymarket market predicts a 34% chance of 2026 being the hottest year on record, while Zeke Hausfather of Berkeley Earth—using data from the European Space Agency's Copernicus programme—projects a 28% chance. This proximity suggests that for certain high-visibility trends, market sentiment can reasonably track expert estimates.

Quantum computing forecasts reveal a different gap. Polymarket gives a 16% chance of a quantum computer breaking Bitcoin by the end of 2027. While recent breakthroughs have accelerated the timeline, theoretical computer scientist Scott Aaronson of the University of Texas notes that researchers still need to assemble systems at a scale necessary to threaten modern cryptography.

These discrepancies highlight that while prediction markets can serve as useful supplements to gauge how scientific information is received by the public, they cannot yet replace the specialized tools and deep expertise required for high-stakes scientific forecasting.

Source: Can Polymarket predict the progress of science, or are subject-experts better?
Domain: nature.com