Hi,

I think it's more easier for for me to explain/ask things with and example. So let's pretend that I'm in the process of designing a MA crossover system with a slower MA and a faster MA.

It's pretty obvious that both of the MA paramters should be optimized and trained.

However, when deciding how to exit the positions, you could do this in different ways. For example, an opposite crossover could place and exit order. An other way to do it is of course to use stop, trail, trailslope and so on. The latter is the one I use in most of my strategies. Usually I optimize the exit parameters, but I've not done it in every strategy.

When I think about it, it feels that the exit parameters are obvious targets for otimizing and training. Do you think I'm 'wrong'? What are your thoughts about this?