If you search through the Collective2 database looking for trading systems, you will always find some trading systems that have performance that seems too excellent.

This is a common problem - and one that affects the entire trading world - not just Collective2. If you assume that all trading system performance is essentially random - and that some will be good and some will be bad - then you will always be able to find the "good" systems... after the fact. The trick, of course, is to find the good systems before they post excellent performance!

Many people have written about this phenomenon. (A great book, if it hasn't been mentioned on this forum already, is Taleb's "Fooled by Randomness" - which may very well change your world view about trading.) As I said, this problem exists also in other trading venues, like the world of mutual funds, where a mutual fund vendor can always offer a fund that has had excellent performance in the past. All the vendor needs to do is make sure he keeps a stable of a dozen funds at all times, and then he can "fold" bad performers into excellent performers, and market only the good ones. Caveat emptor, indeed.

Anyway, I mention this not because I encourage us all to throw up our hands in disgust and walk away from the world of trading systems, but rather, to encourage people to make educated decsisions about the systems they use or subscribe to. The best thing you can do is have realistic expectations about what trading systems can achieve. The reality is that it is simply not possible to earn, say, more than 20% per year without large drawdowns and risk.