Meta-reply: '[C]orrect predictions [someone] made before' = data. Instead of discounting it, I would rather count it more heavily, esp. in a complex domain wh

Meta-reply: '[C]orrect predictions [someone] made before' = data. Instead of discounting it, I would rather count it more heavily, esp. in a complex domain where a full principled evaluation of the evidence is beyond most people. There's a reason you can't read a review paper and then get a PhD in whatever topic. Having to establish everything in a logical chain from first principles every time you want to opine is inefficient, to say the least. Previous performance matters. It's not all that matters, but nobody is claiming that.