#预测市场 Seeing Kalshi's research data, there are several noteworthy details.

Prediction markets have an average error in CPI forecasts that is 40% lower than consensus forecasts. The logic behind this number is quite clear—market participants make decisions based on real economic incentives, and the flow of information and pricing mechanisms are naturally more efficient than unidirectional analyst expectations. More interestingly, when actual data deviates significantly from expectations, prediction markets outperform consensus expectations by 67%, indicating that markets indeed have an advantage in responding to black swan events with faster reaction times and better adaptability.

The 25-month sample period spans a highly volatile economic environment from 2023 to 2025, which enhances the reference value of the data. For institutional research and investment, the signaling value of such prediction markets is becoming more prominent—not as a replacement for traditional analysis, but as a supplementary tool to help make more precise decisions during periods of high uncertainty.

Future points to watch include whether more traditional financial institutions will gradually incorporate prediction market data into their decision-making frameworks, and whether these platforms will expand to forecast more asset classes in the future.