The program I am writing uses technical signals to generates exit points. Since it does not use hard stops, how do I determine risk? My experience with hard stops is that they depend on market conditions. If you have just a fixed dollar amount or percentage, you invariably exit at the wrong time.
The word you are looking for is "assessing". If you trade for a long enough period of time, you will buy and sell at tops and bottoms. It's more important to be confident in your method than to be worried about "appearing" foolish for doing a bad thing at a bad time.
I don't know if it has to do with looking foolish. What I mean is that, for example, you have a stop loss at a particular support. But the stock does not immediately break it or bounce off it. The stock may test the support several times or it may break it but fail to follow through. According to my program, it does not make sense to get out and get back in. The only thing that you will achieve is churning. However, if you just hold and see what will happen, what is the purpose of a hard stop?
It protects you from a potentially big loss, according to the guidelines/rules that you have created for yourself. Focus more on trading strongly trending markets if you're worried about being chopped around.
If you have a system that does not work well with stops, then you don't have a system. What you have is an entry signal that will eventually ruin you. Knowing when and how to exit a trade is as important, if not more so, then knowing how to enter. Your testing may tell you that there is a 90% probability of success with a profit target and no stops. But it only takes one trade gone bad (and it will) to take you to the cleaners. Nobody, and no system, is immune to serious losses without hard stops of some kind.