Today I updated the strategy such that it is possible to check the slow MA and fast MA gradients for a certain angle. I ran these through the optimizer and of course can find combinations of inputs that make money.
However, it seems that the safest strategies are the ones that result in a very small number of trades over the 2010/01-2013/06 period… very small number being 70 trades, with only 30 of them profitable. These do not necessarily always have a high min equity too, so increasing the lot size can lead to margin calls. When the trade count is this low, basically the strategy only follows very strong trends.
This is a good point, optimizing with a lot size that is too small is not realistic enough. I need to use a realistic lot size and account size that will represent my trading. Using realistic values (not the default lot size of 1), a lot of the previously better/winning strategies are margin called.
I need to find some way to reduce the number of losing trades, or limit the losses even more. I haven’t come anywhere near a strategy that I would feel comfortable running in live.