The Sunday Boston Globe's Ideas section has an interview with Philip Tetlock, who has a book on expertise: Expert Political Judgment: How Good Is It? How Can We Know?. Tetlock has asked experts their predictions and then recorded the results. The basic result: experts don't do much better than a coin flip in predicting the future.
Interesting stuff. Experts are very good at reviewing what happened and explaining it through their particular lens of experience and expectations. But when it comes to predicting the future, there are just too many random factors to predict much better than a coin toss. Hopefully, scientific endeavors fare better than this, but we see it a lot.