Welcome back Meta-Traders.It was a blowout week in equities. With September upon us, it was almost like they flipped a switch and stocks ripped higher gaining almost 5% for the week and taking my equity portfolio out to a new 2-year high.
Winners on the week were Apple, Netflix, HMS Systems and Compellent. Okay, i'm clearly on the wrong blog, but it makes me wonder why I spend so much time on Forex when I could be making serious money with stocks. Okay, back to Forex.
It was another quiet week in the Forex market, with profits coming in low and slow. But every account that was active took home some pips and you have to be grateful for that.
The winner on the week was Atinalla #1 which closed a 64-pip winner. Atinalla #1 went into the weekend long EUR/USD and in the black about 125 pips. This trade is shows an open profit of about 3.5% and if closed would bring this account back to down about 5% since open. Keep in mind that this system is a combination of 3 trend-following systems and does well in trending markets. August has been a bit on the choppy side and we expect that this account will make most of its money in just a few trades - which is typically behavior of trend-following systems.
Next up was Megadroid Demo which closed about up about 3.2% for the week. The lot size was doubled since its recovering from a recent drawdown.
For my live accounts, Megadroid Live picked up 5 pips and Scalper X gained 4 pips. While unspectacular, these moves were enough to push both accounts out to new all-time equity highs.
These systems are relatively easy to trade since they have a low "Pain Index" which is a measure of the amount of time spent in drawdown. In case you missed it, Daniel Fernandez did a post on this topic here Pain Index.
I don't necessarily agree with his conclusions that scalping systems are dangerous. He might be right of course, but so far my back-testing, demo testing and live testing don't show that to be the case. Will I be proven wrong? Keep coming back and we'll find out!
Next up was System Q which picked up 29 pips on the week, but is still down about -16% since starting the demo account back on 8/1/2010.
Finally, God's Gift EUR/USD and GBP/USD didn't trade at all this past week. I can't find any problems with the Meta-Trader terminal. Things haven't been quite right since FXDD switched to being a 5-digit broker. If this keeps up, I may have to kill those demo accounts and start up again with God's Gift ATR which is Daniel's adaptation.
One more thing - I started work on an EA which is based on the behavior of the Fish Forex Robot which has been gaining popularity and showing some good results. The system trades EUR/USD starting at 3PM EST 4 days a week. That's all i'm going to say about it for now.
Have a great holiday weekend all!
I hear you on the 5 digit broker FXDD decided to switch too. Luckily, the live account did not go to the 5 digits, so it kept some of the panic upgrading my EA's to a minimum. On the rough side I have been experiencing some tough problems back testing. The Gods Gift 7c has been a problem and I have now just left it alone again. I tried back testing the Gods Gift ATR and had similar problems. It is kind of tough when your back test says 90% modeling quality and you have been told the EA is suppose to average around 50% with the various draw downs of course, and you find you only get 17% return over 10 years back test(something can't be right). My Mover3 EA had some results a few days back then yesterday when I did back testing again the data was not the same. The profitability was not as good anymore, however on a very good plus side my horrible 2010 results are now at -2.4% for the year and not -23%. I think some of what may have happened with my testing is I had stale data at the 4 digit level. I recently deleted/cleaned out my data and re-downloaded 1M time frame data... Tiger Tail experienced some profitability loss in 10 back test as well (I still predict a small average gain of 20% to 30% a year though I am now concerned about using it on GBP/USD). I think a lot of differences has to do with calculations done and it being now rounded to the more significant 4 digits which changes things.
ReplyDeleteNow that I have the data refreshed I need to try back testing the Gods Gift again. I can't help but feel I am just missing something simple here with it, I expect I will figure it out soon. I have been running the Gods Gift 7c and I actually got a trade this week for +15 pips on the EUR/USD pair. The trade happened around 9/2 at about 2:00-3:00. (I kind of can now see where this EA is trying to take pips, very interesting...)
At any rate lots of fun/success'/disappointments etc.
Have a good Labor Day weekend.
JT
JT-
ReplyDeleteThanks for the comment.
I went back and reloaded 5-digit history for GG 0.7c for the M1 and H1 timeframes. Even with that, I can't reproduce my own demo-test results for GG 0.7c for 2010, both accounts loose money.
Looking at the journal tab, i'm seeing a lot of OrderModify Error 130 which means the prices are too close to the market. This means the trailing stop mechanism is not working as intended and that can definitely mess up profitability.
Even before this, I've had a lot of problems back-testing God's Gift, which is why I never answered your question in terms of which was better 0.7c or the ATR version.
So for now, i'm going to let the demo accounts ride. Since I have 9 months of demo testing behind me, I might as well finish the whole year.
I'll probably start a new demo of GG ATR trading EUR/USD and GBP/USD on a single 2000 account at Forex.com. That's what i'm likely to do live starting in 2011.
Thanks again for the comment!
Chris
I kind of suspect that for some reason in back testing the stop move functionality did not seem to work. At any rate, I can at least see when the trades are taking place and start to understand how it works. I plan to forward test for the rest of the year. I need to turn on the ATR version and test that one as well. I wonder how well MT5 does in its back testing, seems that MT4 really has some flaws.
ReplyDeleteJT
Hello Guys,
ReplyDeleteThank you very much for the update Chris :o) As always it is great to see how your trading is evolving week after week.
Regarding the God's Gift ATR and 7c versions, the implementations currently available are actually pretty poor from a coding standpoint reason why errors and difficulty backtesting are common. This week I will be releasing a new version of this EA which will use a much more professional functionally decomposed framework with a great UI and error handling ability. This will be released as part of a global update in which all Asirikuy systems -especially all those that do not use functional decomposition yet- will be changed to a standardized coding framework.
Also JT we are currently implementing Asirikuy systems to run on MQL5 and we will probably have all of them there in about 6-12 months :o).
As always keep up the great work with your blogs guys !
Best Regards,
Daniel