This brings back memories when I was writing code in EasyLanguage for TradeStation. This was not on a professional level I assure, it was for me mostly. Back then forums like these were not as widespread so I was pretty much on my own. I was constantly making changes to MA's or RSI periods, not optimizing; I was looking for something that worked all the time no matter what. Incidently, as I recall I was working with intra-day data on the SP500, similar to what we use when trading FX. The main thing I found out was that the simpler I kept things, the better the result (backtesting) was. I was mainly using indicator conditions because I wasn't too 'good' at coding price action. But, if I can remember, I believe I stumbled on to something that only produced about 2 or 3 trades a week but the results were very good, results being win:loss ratio. This was on the SP500 ($500 a point), the dollar value return was just as good. But this was when the margin to trade the SP500 was something like $25,000. Uh, I was poor then and I'm poor now. That was more of a hobby and I had and still have that full time job away from home thing. EA's were not 'available' 10 years ago or if they were the real-time data and ability to trade that way in the first place was $$$$$. The point 'I'm trying to get to' is that I kinda burned myself out looking for that one thing that would fit, say, on a corn contract. Nope, what works for one market doesn't necessarily work on another. Different markets have different behaviors. Whether your trading something that has roots and hooves or makes a ching-ching sound in your pocket, they are different. Now, put a corn chart and a EUR/USD chart next to each other without any legends and they look pretty damn similar don't they? But, still they behave differently. I can't answer why the same EA shouldn't produce 'similar' results if the charts looks 'similar', they just don't. It was and still is very frustrating to write code (EA's) for hours on end and come up with, not nothing, because you always learn from doing. But, you do learn what not to do. Writing systems did teach me that just because an indicator is overbought(sold) that the market will turn the other way. Markets can stay overbought(sold) for a long time but indicators don't care. They are all formulated from price action. So, might that be the key??? Price action??? I think it's a lot more difficult to program the price action of a market (behavior) vs. 'IF RSI > 70 THEN SELL'. I applaud Hendrick for working on 'the project' and I doubt very seriously he came away with nothing. He learned something I gaurantee you that. Open up that bottle of red Hendrick and relax for a while. The markets will still be there when you wake up from your slumber. Hope to hear back from you soon.
Good Trading To All
Good Trading To All