Prediction Market Alpha: Where Markets Are Wrong

The Opportunity

Prediction markets aggregate crowd wisdom. But crowds can be wrong. Finding systematic errors = alpha.

Research Questions

1. Where do markets misprice?

   • Recency bias (overweighting recent events)
   • Narrative capture (what sounds true vs is true)
   • Liquidity gaps (thin markets = noise)
   • Time horizon confusion

2. What signals predict outcomes better than markets?

   • Expert opinions vs crowd
   • Base rates vs market odds
   • Leading indicators

3. How to calibrate predictions?

   • Brier scores
   • Tracking methodology
   • Adjusting for overconfidence

Data Sources

• Polymarket (crypto, politics)
• Metaculus (tech, science)
• Kalshi (events, economics)
• Manifold (everything)

Experiments

1. Track 20 predictions for 30 days
2. Calculate calibration
3. Identify systematic biases
4. Develop contrarian signals

Output

Weekly Alpha Brief with market mispricings.