There are two things we know for certain about past performance: First, past performance is no guarantee of future performance. That's about the only thing Wall Street will tell you that you can take to the bank. And they are forced to tell you that. Second, past performance attracts lots of new money. The longer the string of performance lasts, the more money pours into the fund. And then the string runs out leaving the new arrivals to wonder what happened. How could this hero have lost his mojo? Perhaps he never had any in the first place. The record if full of stars that imploded costing investors billions as their funds got sucked into a black hole. I cite Bill Miller as one of the latest wonder boys banished in ignominy. Peter Lynch is one star that bailed out at the top of his fame. He grew Fidelity's Magellan Fund from obscurity to the largest mutual fund on the planet. After his retirement he stayed on at Fidelity to nurture and mentor Fidelity's aspiring fund managers. If there was anybody who could pick a good fund manager it would be Peter Lynch. Yet the abysmal performance of Magellan Fund in the interim is mute testimony to the impossibility of picking winners in advance. Across the entire universe of funds, active management costs investors from one to two percent compounded per year. We should expect under-performance. It would be bad enough if the opportunity cost was a steady one to two percent per year. But, it gets worse. Active management adds another layer of risk on top of market risk. There is a wide variation around the index return which at the tail can be devastating. The distribution is actually far worse than it looks. The funds listed were the survivors and Morningstar doesn't include the several hundred funds that failed along the way. These failed funds normally get merged into other funds with better records making their performance magically disappear from the results. Survivor bias badly skews the illustration. I grew up believing that managers added value, so it took a lot of evidence to change my view. But, literally hundreds of studies indicate it's not so. Not one credible shred of evidence supports active managers. I believe that the case has been proved beyond a shadow of a doubt, and even higher standard than necessary in a criminal case.