I started partecipating, but got turned off by the ridicolously detailed questions outside my area of expertise. Do I think a sack of rice will fall over when the Ethiopian delegation visits Ecuador in March? How sure am I about my prediction? It doesn't seem to help me to achieve better calibration. I'm curious if people that are partecipating are getting value out of it, and what kind of value.
Previously: "Test Your Forecasting Ability, Contribute to the Science of Human Judgment" (May 2012), "Get Paid to Train Your Rationality" (August 2011)
-
Think you have what it takes to make good predictions? Since 2011, the Good Judgment Project (GJP) has been making predictions on issues of international relations and foreign affairs, recently winning the IARPA (Intelligence Advanced Research Projects Activity) prediction contest. Predictions from the GJP have been startlingly accurate, outperforming prediction markets, and exceeding even optimistic expectations. It's run by Phillip Tetlock, the famous predictor of "foxes and hedgehogs" fame.
From the Monkey Cage article:
But the good news is that you now have a chance to get involved with GJP Season 3 if you think you're a great predictor:
-
Also cross-posted on my blog.