The second thing is that you probably should not put so much store in backtest. They do not need to be as accurate as you seem to want for a very simple reason.
The only real/valid use of a backtest is to answer the question: "Given perfect knowledge of the future, would my system have made money?". In other words, you use a backtest to decide if you need to do further investigation towards using the system, because if the system cannot make money in backtest, there is a probability, bordering on certainty, that the system cannot make money in real time.
Given that aegis, what does it really matter if a few trades are not taken because the market does not retreat. If that is the case, that is what would have happened in reality anyway, so that is all to the good. Of course, you can switch to Market Orders, but in fact, in backtest, Market Orders will probably underestimate slippage, relative to reality, and will actually give you results which are likely more deceptive.
Leave a comment: