I think the point of the agreement theorem is that it has nothing to do with real people—it shows how far everyone is from Bayesianity almost all the time.
I think the point of the agreement theorem is that it has nothing to do with real people—it shows how far everyone is from Bayesianity almost all the time.