There were 69 days left in the forecast period. So David divided 69 by 365. Then he multiplied by 1.8. Result: 0.34. So he concluded that there is a 34% chance the answer to IARPA’s question would be yes.
Example of how superforecasters make their calculations - the actual math of it