Super Carry Trade Trading System | Page 8 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Carry trade vs carry trader 3 replies

Super Carry Trading Strategy - Expert Advisor 22 replies

long term carry trade system? 3 replies

Gamblers carry trade system 5 replies

Attachments: Super Carry Trade Trading System
Exit Attachments

Super Carry Trade Trading System

  • #141
  • Feb 22, 2008 1:44pm Feb 22, 2008 1:44pm
Quote
Correct me if I am wrong.. the average price would be 211.02 (210.97+211.07 = 422.04 / 2 = 211.02). The next ideal new base trade would be at 209.02 or below. Right?

This the way I understand it as well. If 209.02 is filled, then close out all trades when open P/L equals 0 and then start new series of trades.
  • #142
  • Feb 22, 2008 4:20pm Feb 22, 2008 4:20pm
The, usually 2, sometimes 1 stranded positions is the cost of collecting pips on the way up to wherever the market stops. If you want to make sure you only have one base position that might get stranded you should consider Timefreedom's Aphid EA method.

Now it sounds to me like you are planning on entering an averaged down position at exactly 200 pips below your average. That may be a viable method, but as I understand it, that is more like an optional way of doing things. In the system that gave birth to the one we talk about here, the system's founder says he checks price/positions just once a day, and he isn't real methodical about when he does it. In the Aphid EA, this is a little more structured/codified so that an evaluation is still made only once a day, but it is at a single specific time. In other words, price might go down 500 pips, reverse, and end up being down only 90 pips by the time the review time strikes. Under the Aphid system and the original system, you wouldn't make any adjutment (no average down at all). I don't know how Hayseed is coding this to work in his EA. If it were an option we could backtest and maybe get a feel for whether or not the option for Buy Limit orders at 200 pip increments below the average makes the system better. It might very well be superior, or it might not.

In any case, your option would force potentially more averaging calculations, and new grids throughout the day. This could actually force quite a few more stranded positions on the way down. I mean, if you figure for each 200 point drop you have potentially 2 additional stranded positions as price might do a small retrace; your number of open positions on a 1000 pip retracement could be intolerable -- or possibly even break the bank. I'm not saying it isn't worth looking into though. Just be aware that updating once a day has a way of limiting your risk. If the market fall 500 pips, you're potentially going to have significantly more positions at risk than I will with just 1 new averaged down position 500 pips below my 2 high stranded positions. I think if you run through the math in that situation you'll find the once a day method plays in your favor with regard to what your average price is. If you do the calculation with Buy Limit orders every 200 pips, I think this scenario causes you average price to be higher. That won't always be the case. Sometimes your method puts the average closer to the current price. But the more I think on it, the more I believe your method is advantageous in the smaller or quicker recovering drops. Dreamliner's method is advantageous in the longer drawdowns, which are the ones that will kill your account. In an average market your option should be more profitable... but can we guarantee today and or tomorrow will be "an average" market day. You know, you only have to be wrong on that score once, to be out of the game.
  • #143
  • Feb 22, 2008 9:48pm Feb 22, 2008 9:48pm
still struggling with the advanced ea..... so in an effort to get us all used to what is going on , here is timefreedoms ea with the buystop grids for demo only.... just as a first non trading step toward a finished version....

hopefully everyone understands the risk here .....

couple things..... the spreads will vary, a 10 by 10 grid such as dreamliner suggest might not work with all vendors.... the buystops distances are all setable.... also for obvious reasons you might want the takeprofit settings different at times.... they are setable....

if you think about how a grid is laid out you'll see the default settings equate to an order every 10 pips with a profittarget every 10 pips....

the 'delay' setting is to set the expiration of the buystops..... after 5 days of backtesting on almost every pair, it appears 72 hours is enough....

the setting 'averagevalueincrease' is to bump up the function that closes all orders at breakeven.... it's by default 0, setting it to 10 requires the market to move 10 more pips before the ea closes everything, 20 requires 20 and so on.... if you think about it, you'll see reasoning..... keep in mind, there is a point of diminishing returns so be realistic....

and last, there is a setting saying 'demo only'.... by default it's set to false.... this is to remind us all of the risk ..... set it to true for demo testing.... do not use this in a live trading account....

it's purpose is just to familarize ourselves with the ea to come..... and once again, this thing will place trades so don't use it in a live account....h
Attached Image(s) (click to enlarge)
Click to Enlarge Name: demo only.gif Size: 28 KB
Click to Enlarge Name: demo only 2.gif Size: 32 KB
Click to Enlarge Name: demo only.jpg Size: 228 KB
Attached File(s)
File Type: mq4 demo only.mq4   35 KB | 1,042 downloads
to trade and code, keep both simple... no call to impress....h
  • #144
  • Feb 22, 2008 10:32pm Feb 22, 2008 10:32pm
It seems so silly that people would trust the EA while this strategy takes only ten minutes to set up daily manually. Dump the EA and do it like a man and be proud of your effort.
  • #145
  • Feb 22, 2008 10:52pm Feb 22, 2008 10:52pm
hey fxhound.... pride has nothing to do with it, nor does ego , nor does being a man.....h
to trade and code, keep both simple... no call to impress....h
  • #146
  • Feb 23, 2008 12:23am Feb 23, 2008 12:23am
Good job Hayseed... Please keep up the diligent effort... it's appreciated...

Walt
  • #147
  • Feb 23, 2008 9:23am Feb 23, 2008 9:23am
Thank you for this trading system Dreamliner. What time(s) of day do you personally check your trades?

Thanks,

Glen

P.S. I saw a post where you mentioned being a flight instructor. What aircraft do you teach, the Dreamliner ?
  • #148
  • Feb 23, 2008 11:24am Feb 23, 2008 11:24am
Good day everybody, I'm very new to this thread, and i will be very happy if any of my seniors could put me through. is there an E.A for this trading system? the one i saw gave a margin call result on strategy testing. I would have paper trade the system but i seems not to get when the carry trade begins. or when exactly do i set the orders! Thanks a lot.
  • #149
  • Feb 23, 2008 11:49am Feb 23, 2008 11:49am
Osta, about EAs--see Hayseed's post #90. You can enter the market at any time to begin and check your trades daily any time you want. Review Dreamliner's post #1 for all the rules...it's all there.

Glen
  • #150
  • Feb 23, 2008 7:26pm Feb 23, 2008 7:26pm
Youe efforts and time are greatly appreciated. Will backtest and report my findings.

TimeFreedom

Quoting hayseed
still struggling with the advanced ea..... so in an effort to get us all used to what is going on , here is timefreedoms ea with the buystop grids for demo only.... just as a first non trading step toward a finished version....

hopefully everyone understands the risk here .....

couple things..... the spreads will vary, a 10 by 10 grid such as dreamliner suggest might not work with all vendors.... the buystops distances are all setable.... also for obvious reasons you might want the takeprofit settings different at times.... they are setable....

if you think about how a grid is laid out you'll see the default settings equate to an order every 10 pips with a profittarget every 10 pips....

the 'delay' setting is to set the expiration of the buystops..... after 5 days of backtesting on almost every pair, it appears 72 hours is enough....

the setting 'averagevalueincrease' is to bump up the function that closes all orders at breakeven.... it's by default 0, setting it to 10 requires the market to move 10 more pips before the ea closes everything, 20 requires 20 and so on.... if you think about it, you'll see reasoning..... keep in mind, there is a point of diminishing returns so be realistic....

and last, there is a setting saying 'demo only'.... by default it's set to false.... this is to remind us all of the risk ..... set it to true for demo testing.... do not use this in a live trading account....

it's purpose is just to familarize ourselves with the ea to come..... and once again, this thing will place trades so don't use it in a live account....h
  • #151
  • Feb 24, 2008 7:29am Feb 24, 2008 7:29am
Hayseed, Thank you for your efforts.

You have me very curious now though. This demo EA seems to have all the functionality we have discussed in this forum to date. So, I'm wondering what additional functionality you are seeking to include. I certainly understand if you are too busy working on things to reply, but, now that I've sullied the waters with the question, I'm probably speaking for the masses here.
  • #152
  • Edited 12:20pm Feb 24, 2008 12:06pm | Edited 12:20pm
Hi Hayseed,

I will be testing this on a demo account. It looks very interesting. A friend of mine who programs his own ea's wanted to find someway to auto optimize them. He found this link on the mql website and he indicated it has made his ea's become profitable. Not sure if this will help you or not since I am not a programmer. Here is the link. http://articles.mql4.com/506

Jeff

Quoting hayseed
still struggling with the advanced ea..... so in an effort to get us all used to what is going on , here is timefreedoms ea with the buystop grids for demo only.... just as a first non trading step toward a finished version....

hopefully everyone understands the risk here .....

couple things..... the spreads will vary, a 10 by 10 grid such as dreamliner suggest might not work with all vendors.... the buystops distances are all setable.... also for obvious reasons you might want the takeprofit settings different at times.... they are setable....

if you think about how a grid is laid out you'll see the default settings equate to an order every 10 pips with a profittarget every 10 pips....

the 'delay' setting is to set the expiration of the buystops..... after 5 days of backtesting on almost every pair, it appears 72 hours is enough....

the setting 'averagevalueincrease' is to bump up the function that closes all orders at breakeven.... it's by default 0, setting it to 10 requires the market to move 10 more pips before the ea closes everything, 20 requires 20 and so on.... if you think about it, you'll see reasoning..... keep in mind, there is a point of diminishing returns so be realistic....

and last, there is a setting saying 'demo only'.... by default it's set to false.... this is to remind us all of the risk ..... set it to true for demo testing.... do not use this in a live trading account....

it's purpose is just to familarize ourselves with the ea to come..... and once again, this thing will place trades so don't use it in a live account....h
  • #153
  • Feb 24, 2008 2:46pm Feb 24, 2008 2:46pm
hey azjeff.... thanks for the link , i'll take a peak.....

hey qman..... perhaps it was me that sullied the waters by offering to write an ea when none was requested.... maybe we should start a new thread to further investigate timefreedoms ea..... just as thought.....

the " additional functionality " is along the line of drawdown protection but more to the point, drawdown exploitation.....

as seen below, 1k went to 14k in the same 2 years.... this is timefreedoms ea with dreamliners buystops along with the added exploitation code.....

there are a large pecentage of trades placed but not filled, far too large a number..... it's those never-to-be-filled orders that are causing the trouble.....

we might also look into the number of sunday troubles filled in any gap up..... if the market ever gapped up only to fade back, there might be many buystops filled against us...... not sure.....

we also need to limit the lots, once this thing gets a good grip it could go parabolic quickly.... that's a negative..... h
Attached Image (click to enlarge)
Click to Enlarge Name: demo only.gif Size: 28 KB
to trade and code, keep both simple... no call to impress....h
  • #154
  • Feb 24, 2008 5:39pm Feb 24, 2008 5:39pm
Hayseed, okay, I see where you are going with this. You've not explained what the method is for the drawdown exploitation, but you are probably right -- this may no longer be the right place.

Dreamliner, I think this becomes your call at this point. Development beyond what is already here is functionality beyond your original concept. I gather you are not especially interested in an EA beyond what is working so far. If we continue with this here we will be hijacking your thread. Out of deep respect for what you have shared, I wouldn't want to do that.

But Hayseed may be onto something even bigger... To find out, we will be talking beyond your original idea. So, do let us know if you want this to continue here under your name, or if you prefer to keep your original plan to remain pristine.

Quoting hayseed
hey azjeff.... thanks for the link , i'll take a peak.....

hey qman..... perhaps it was me that sullied the waters by offering to write an ea when none was requested.... maybe we should start a new thread to further investigate timefreedoms ea..... just as thought.....

the " additional functionality " is along the line of drawdown protection but more to the point, drawdown exploitation.....

as seen below, 1k went to 14k in the same 2 years.... this is timefreedoms ea with dreamliners buystops along with the added exploitation code.....

there are a large pecentage of trades placed but not filled, far too large a number..... it's those never-to-be-filled orders that are causing the trouble.....

we might also look into the number of sunday troubles filled in any gap up..... if the market ever gapped up only to fade back, there might be many buystops filled against us...... not sure.....

we also need to limit the lots, once this thing gets a good grip it could go parabolic quickly.... that's a negative..... h
  • #155
  • Feb 24, 2008 9:00pm Feb 24, 2008 9:00pm
Thanks for sharing this great link. "The main idea is to start optimization of an EA automatically once a day and then analyze the obtained optimization results and record them in the EA's variables". I will do some homework and try and optimize Aphids variable: AddLotsThreshold, which is fixed to -100 pips right now. Dreamliner currently uses -200 pips to add additional lots when price is retracing and I use -100 pips. Maybe -100 pips is best or maybe -200 pips is better. I will run this automated optimizer everyday and I will evaluate what number it suggests for the AddLotsThreshold.

TimeFreedom

Quoting azjeff
Hi Hayseed,

I will be testing this on a demo account. It looks very interesting. A friend of mine who programs his own ea's wanted to find someway to auto optimize them. He found this link on the mql website and he indicated it has made his ea's become profitable. Not sure if this will help you or not since I am not a programmer. Here is the link. http://articles.mql4.com/506

Jeff
  • #156
  • Feb 24, 2008 9:05pm Feb 24, 2008 9:05pm
If Dreamliner decides he wants to keep this thread more about his strategy, which is totally understandable, then please count me in to discuss new idea's to optimize Aphids results in a separate thread.

TimeFreedom

Quoting Qman
Hayseed, okay, I see where you are going with this. You've not explained what the method is for the drawdown exploitation, but you are probably right -- this may no longer be the right place.

Dreamliner, I think this becomes your call at this point. Development beyond what is already here is functionality beyond your original concept. I gather you are not especially interested in an EA beyond what is working so far. If we continue with this here we will be hijacking your thread. Out of deep respect for what you have shared, I wouldn't want to do that.

But Hayseed may be onto something even bigger... To find out, we will be talking beyond your original idea. So, do let us know if you want this to continue here under your name, or if you prefer to keep your original plan to remain pristine.
  • #157
  • Feb 24, 2008 9:54pm Feb 24, 2008 9:54pm
I think this would be a good use for the optomization routine. But I believe we should be shooting for, not is -100 or -200 better... I think it should be more along the lines of, is the current market environment better suited to the use of -75, and tomorrow, will we find that -145 is a better, and next week, when things are rocketing away from us, maybe the optimization will tell us to not add a position until we are -400. That would be taking the power of this to the next level -- along with Hayseeds thought of some trigger to actually (possibly) step into the trend running away from our original position --- Still holding out on that until I know more about the logic in the trigger, and the method of the money and position management.

In the end, I'd be really happy knowing that an EA was going to be wise enough to realize the market has fundamentally changed, at least in the near term; and decides it must adjust accordingly. That should effectively minimize our steps in against the trend, and thus lower our average for closing all the averaged positions when we do, as well as reduce the number of stranded positions we have to average (which all boils down to less risk in the system, so I might be able to increase my percentage and my eventual winnings). --- NOT that there is anything wrong with what Dreamliner has given to us. That really looks like a winner to me, and I am in fact collecting pips using it since last Friday.

Quoting TimeFreedom
Thanks for sharing this great link. "The main idea is to start optimization of an EA automatically once a day and then analyze the obtained optimization results and record them in the EA's variables". I will do some homework and try and optimize Aphids variable: AddLotsThreshold, which is fixed to -100 pips right now. Dreamliner currently uses -200 pips to add additional lots when price is retracing and I use -100 pips. Maybe -100 pips is best or maybe -200 pips is better. I will run this automated optimizer everyday and I will evaluate what number it suggests for the AddLotsThreshold.

TimeFreedom
  • #158
  • Feb 25, 2008 6:05am Feb 25, 2008 6:05am
Greetings friends,

I'm happy to watch this thread develop, and so far all I've seen are your great improvements, which is what we are all after. So hijack away, just don't demand a ransom .
  • #159
  • Feb 25, 2008 6:07am Feb 25, 2008 6:07am
Haysee, I think the looks of this equity curve are amazing. But can you explain why orders placed but never filled are a problem? That shouldn't seem to have anything to do with equity curve or anything else, should it?

Quoting hayseed
hey azjeff.... thanks for the link , i'll take a peak.....

hey qman..... perhaps it was me that sullied the waters by offering to write an ea when none was requested.... maybe we should start a new thread to further investigate timefreedoms ea..... just as thought.....

the " additional functionality " is along the line of drawdown protection but more to the point, drawdown exploitation.....

as seen below, 1k went to 14k in the same 2 years.... this is timefreedoms ea with dreamliners buystops along with the added exploitation code.....

there are a large pecentage of trades placed but not filled, far too large a number..... it's those never-to-be-filled orders that are causing the trouble.....

we might also look into the number of sunday troubles filled in any gap up..... if the market ever gapped up only to fade back, there might be many buystops filled against us...... not sure.....

we also need to limit the lots, once this thing gets a good grip it could go parabolic quickly.... that's a negative..... h
  • #160
  • Feb 25, 2008 7:36am Feb 25, 2008 7:36am
hey dreamliner.... something is causing occasional orders not to be closed but dragged along for months..... timefreedoms ea does not do this , so it must be something in my added code.... possibly in the way i'm counting the orders........


that equity curve is being accomplished by the ea have more to work with than just a time of day.... the orders are being placed with the trend and very few placed for any reason againt the trend..... this 'somewhat' eliminates the need for averaging down....

so the ea has more control over that -100 or -200 entry level.... sometimes it might place orders at -50 and other times it might not place any orders till -300.....

also, 10 pips for a profit target is kinda tight..... if the ea chooses the profit levels using something like atr, it should payout better.....

as seen in the chart below, many of those orders that dragged down the account at the end of the test were months old..... hundreds of other orders have come and gone but these remaining orders are escaping all attemts to close them....

not sure why..... still studying on it....

this test here is not using the drawdown protection code, so the number of orders is about half..... ie, it's long only....h
Attached Image (click to enlarge)
Click to Enlarge Name: aphidsgbpjpy.gif Size: 28 KB
to trade and code, keep both simple... no call to impress....h
0 traders viewing now