We used Robin Hanson's logarithmic market scoring rule (http://hanson.gmu.edu/mktscore.pdf http://hanson.gmu.edu/mktscore.pdf). The combinatorial aspect allowed

We used Robin Hanson's logarithmic market scoring rule (http://hanson.gmu.edu/mktscore.pdf http://hanson.gmu.edu/mktscore.pdf). The combinatorial aspect allowed us to try to answer questions about the correlation between different securities in the market - hopefully more interesting for policy analysis than a simple predictor of the future. The main securities in the market were 5 statistics that we were tracking quarterly in 8 different Mid-East countries (things like civil stability, military activity, US aid, etc). Then there were supposed to be a handful of one-off type events; we hadn't decided on them yet, but gave some "ripped from the headlines" examples that were the main focus of the criticism. The values of all these securities were to be judged by the Economist Intelligence Unit.