I wouldn't classify it quite the same way. But essentially yes, I am still employing my strategy "data mining" approach.
I guess I'm curious why the performance was so incredibly smooth at one point. Whatever those market conditions were... I'm wondering what your model missed to figure out why the right conditions weren't present to the right of that smooth piece of the equity curve.
No strategy is going to be right all the time.I suspect there are periods where my strategy will do extremely well.. Others it will go through a rough patch. As long as it eventually makes a new high (which I think it will soon), then there is nothing to worry about.
No, I meant, what do you think the model missed or was unable to dodge with the less successful trades?
That is the million dollar question. If I knew the answer, I would be rich. There are two general observations I have made: Observation#1: NT + MBTrading suffers from what I call "phantom trades". I have noticed that during the "smooth" periods, I have very few phantom trades. Sometimes even going a week between instances. When I have more phantom trades,my results become choppy. Part of the choppiness is due to "incorrect trades", but I wonder if there is something else at work. I notice phantom trades increase during times where I get a lot of disconnects from MBTrading. Perhaps something to do with the data causes my strategies to under perform during these periods. Observation #2: While I am trading 5 currency pairs both long and short utilizing 2 different time frames for a total of 10 strategies, I seem to maintain a bit of correlation with the EURUSD. During periods where EURUSD is rising hard,the strategy does well. When choppy, the strategy chops as well. This correlation seems to be fading however.
Its not so simple. I am trading near the upper end of what I deem to be acceptable leverage for my strategy. If I were to change to IB for example, there is a minimum commission of $2.50 per trade. In order for commission to "make sense",it would require an account value of about $40,000 at current leverage levels. I don't have enough to fund that size account. Therefore, I have to stick it out where I am and look for ways to minimize the phantom trades (which I have mostly done)
You might find the higher commission in offset by the tighter spread. I mean .. in the forex world, you'll be getting screwed about the same amount wherever you go. Largely a question of whether it's by explicit fees, or spread imho.
It sounds like Frostengine's problems is a simple technical one that can be fixed with a more competent programmer. Also, remember that as your leverage/pip value increases with IB, the $2.50 commission becomes practically irrelevant as your notional amount increases above $100,000 - $200,000+. Therefore, the "commission problem" will be a short-term one if the strategy can continue to perform.