Trading using Random Entries (Van Tharp book method) - Traders' Roundtable (
At the above link, you can find some pretty comprehensive testing and discussion of the Van Tharp random entry method.
One of the inherent problems with the random entry system is that it is difficult to get a positive expectancy (really hate that word). You would always trade with stops, but your exit criteria may be different. For instance, you could exit either using a trailing stop or a fixed profit target. With trailing stops the size of your winners relative to your stops becomes smaller. Thus, you need the win% to be higher, but it probably won't happen in practice. Using targets, you may think that a 33% win percentage is possible, but in practice, I found that larger targets reduced my win%.
In summary, trying to increase my win% reduced the size of my winners to such an extent that the system was no longer profitable. Trying to increase the size of my winners reduced the win% so that the system was no longer profitable.
PS. My tests were not done with TradingBlox and all of the futures markets, but in Excel with 6 major currency pairs. Entries were time-based, for instance if there is no position, take a random position anywhere between 1-10 days after the last trade was exited.