FutureEval Spring Results: Pros Beat Bots, but the Gap is Nearly Gone by Benjamin Wilson 🔸

Main TakeawaysTop Find­ings:Pro fore­cast­ers beat bots but with­out sig­nifi­cance: Our team of 10 Me­tac­u­lus Pro Fore­cast­ers out­performed the top-10 bot team by an av­er­age of 1.25 head-to-head points per ques­tion. But this edge is small rel­a­tive to the ques­tion-to-ques­tion vari­a­tion and not statis­ti­cally sig­nifi­cant (one-sided p = 0.247), so this sam­ple doesn’t give us enough ev­i­dence to con­firm Pros were gen­uinely ahead this sea­son.The bot team im­proved no­tably in th...