>If your predictions consistently fail, you're not doing good science. I agree with where you're going, but I think the better phrasing would be "If your predi

>If your predictions consistently fail, you're not doing good science. I agree with where you're going, but I think the better phrasing would be "If your predictions consistently fail, your scientific theory isn't very good" Good science usually involves a lot of failure before it provides good theory. This doesn't excuse incredible overselling of economic models by economists though.