Minds Almost Meeting Polls vs. prediction markets (Robin Hanson & Agnes Callard, with Arnold Brooks)
Oct 31, 2024
In this engaging conversation, Arnold Brooks, an expert in prediction markets, joins economist Robin Hanson and philosopher Agnes Callard. They dive into the intriguing clash between polls and prediction markets, unpacking how betting dynamics can influence forecasting accuracy. The discussion touches on the roles of informed and uninformed traders in these markets and the societal value derived from structured betting. They also reflect on how ancient oracle practices compare to modern decision-making tools, showcasing the shift toward crowd-sourced wisdom.
AI Snips
Chapters
Transcript
Episode notes
Forecasting Stages
- Polling, statistical analysis, and prediction markets are distinct stages of forecasting, like ingredients, preparation, and the final dish.
- Don't treat them as directly comparable information sources.
Value of Closeness Information
- Knowing an election is close might motivate voting, but revealing closeness can also discourage voting if it's not close.
- The value of information depends on averaging over all possible information states.
Wolves and Sheep
- Financial markets function with 'wolves' (informed traders) and 'sheep' (noise traders).
- Sheep, while adding noise individually, create trading opportunities that attract wolves and improve market accuracy.