Going Home - a multi-pair auto-trader for Dreamliner's strategy, by Steve Hopwood | Page 5 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

building multi-pair multi-timeframe scanner 48 replies

multi-pair percentage scale charts 28 replies

Price info indicator needed - one pair's price x another pair 5 replies

Multi-pair trading or Few-Pair Trading. [poll] 15 replies

Need help modifying Multi Pair Stochastic indicator 0 replies

Attachments: Going Home - a multi-pair auto-trader for Dreamliner's strategy, by Steve Hopwood
Exit Attachments

Going Home - a multi-pair auto-trader for Dreamliner's strategy, by Steve Hopwood

  • #81
  • Feb 1, 2011 5:42am Feb 1, 2011 5:42am
Quoting SteveHopwood
Thanks for trying the 30.

Any takers for 20?

I have 20

I took away all CAD. They move more sideways than up and down. Pinched more than half of my profit one hour ago.
  • #82
  • Feb 1, 2011 5:58am Feb 1, 2011 5:58am
Quoting E2u
I have 20
Great.

Quote
I took away all CAD. They move more sideways than up and down. Pinched more than half of my profit one hour ago.
Now we know the basic code is working, I can start to expand. I shall make the pairs to trade an input, so people can easily remove them without having to go into the code.

  • #83
  • Edited 6:30am Feb 1, 2011 6:13am | Edited 6:30am
http://www.myfxbook.com/widget?id=81...=1&color=green

This is a new account with a -0.09 test trade only as of 01 feb 2011 1.15pm CET
Any testing you want me to do I can
Let me know and I will just change settings to suit
Really have to watch news with this thing
  • #84
  • Edited 7:21am Feb 1, 2011 7:15am | Edited 7:21am
Sorry sorted now
  • #85
  • Feb 1, 2011 7:37am Feb 1, 2011 7:37am
The latest and final (for now at least) update is in post 1.

I have added a PairsToTrade input so users can easily remove pairs they do not like or which they know to be unsuccessful - E2u recommends removing CAD pairs.

There is an input to make GH calculate differential lot sizes based on each pair's tick value. The difference might be too dramatic if you are trading tiny lot sizes.

Both appear to have worked properly on the latest gbpchf/gbpaud trades taken.

______________________________________________________________

A few words of caution here. We have already had Matt telling us that this will not work long-term. Those of us familiar with his work know how knowlegable and experienced he is.

I quote here a section of a pm sent to me by astral, holding a reply from scoobs in answer to a query about the validity of this trading method (I know neither of these two individuals will object to me quoting them):
"
Originally Posted by scooby-doo
Correlation trading will never work long term. There is no relationship between Stoch and a symbols price. Yes, they will make short term gains but sooner or later they will end up negative overall. If it worked then dont you think the big boys would just use 1000000 contract size and scalp the market to death ?."


For newcomers, scooby-doo is a professional floor trader with many years experience.

These individual views are well worthwhile bearing in mind if GH appears to be successful initially. All we can do is test it and see.

  • #86
  • Feb 1, 2011 7:41am Feb 1, 2011 7:41am
Whats interesting to note is that because I am naturally slower with my manual method , I tend to be getting bigger profits on them sometimes
Whereas the bot triggers at 80 , if I dont have squalou's ea on line ; by the time I have it set up its is at a peak maybe I suppose
Also as squalou points out on the other thread spikes are not shown on the Stoch , for example I have GBPAUD GBPCHF open but the stoch does not show at anytime near to 80 %

Good idea re taking out certain currencies
  • #87
  • Feb 1, 2011 8:56am Feb 1, 2011 8:56am
Quoting peterporter
Whats interesting to note is that because I am naturally slower with my manual method , I tend to be getting bigger profits on them sometimes
Whereas the bot triggers at 80 , if I dont have squalou's ea on line ; by the time I have it set up its is at a peak maybe I suppose
Also as squalou points out on the other thread spikes are not shown on the Stoch , for example I have GBPAUD GBPCHF open but the stoch does not show at anytime near to 80 %

Good idea re taking out certain currencies
Now we know the bot is working, it might be worthwhile experimenting with a higher stoch diff trigger.

I forgot to alter the default maximum spread, so everyone remember to adjust it to a more reasonable figure.

  • #88
  • Feb 1, 2011 11:56am Feb 1, 2011 11:56am
Quoting E2u
I have 20

I took away all CAD. They move more sideways than up and down. Pinched more than half of my profit one hour ago.
E2U
How are your trades panning out at 20 , and are you using 80 as a trigger or higher
  • #89
  • Feb 1, 2011 12:05pm Feb 1, 2011 12:05pm
There is an interesting post by ES Trader
Maybe we could incoporate that sort of strategy within the bot

http://www.forexfactory.com/showpost...postcount=2051

Ta
  • #90
  • Feb 1, 2011 12:28pm Feb 1, 2011 12:28pm
For what its worth I have to agree with scooby-doo.
I suspect when the markets are moving as they are at the moment people will make steady and small profits but then the pairs will re-align themselves and all the previous profits will be blown in one go.
Meanwhile, be lucky........cos that's what you need to be......
He was looking for the card so high and wild hed never need to deal another
  • #91
  • Feb 1, 2011 2:38pm Feb 1, 2011 2:38pm
Quoting peterporter
E2U
How are your trades panning out at 20 , and are you using 80 as a trigger or higher
I have everything default. Lot size 0.10 and CloseTradeStoch =20 and no CAD
It looks like 20 is similar to change money only. Plus Minus Zero
6 order closed with 5 Euro minus tot. I continue to try 20 until tomorrow.
  • #92
  • Feb 1, 2011 2:42pm Feb 1, 2011 2:42pm
Quoting dr_who
For what its worth I have to agree with scooby-doo.
I suspect when the markets are moving as they are at the moment people will make steady and small profits but then the pairs will re-align themselves and all the previous profits will be blown in one go.
Meanwhile, be lucky........cos that's what you need to be......
I have written some indicators for this system, but I have not necessarily endorsed it, but I am a bit confused when it is said that it will not work, simply because if it did, it would have been arbitraged away and you saying that "you suspect" it won't work. I am very interesting in hearing or seeing hard data that shows it will not work. Otherwise there is nothing left to do, but to test it. If you or anyone else has any hard data, I would sure like to see it, not just that it has not been done. That would really be helpful and speed up the process.
  • #93
  • Feb 1, 2011 3:30pm Feb 1, 2011 3:30pm
Quoting smjones
I have written some indicators for this system, but I have not necessarily endorsed it...
Your are correct SIR with regard to testing. One thing I have realized during the short time I have been doing forex is like most other skills, there is no substitute for EXPERIENCE. Reading books, articles, forums, seminars, courses they all help (because you are absorbing the fruits of other people`s experience). But at the end of the day nothing replaces what the retail traders call "Screen Time".That is why I highly value the opinion of people like Macman, steve, Bob and your good self to name a few.

Scoobs is in a different league and we are very fortunate that he freely shares his opinion with us. If I come across a concept that feels very exciting I impose on his time and ask his opinion. His replies are objective and rational and thus far he has been right all the time. I think the reason Steve chose to share the contents of the pms was to ensure people do not come to have unrealistic expectations of this system (i.e. Holy Grail). Having said all of that knowing Scoobs he would be more than happy if tests prove him wrong in the long run.

Kind Regards
  • #94
  • Feb 1, 2011 4:23pm Feb 1, 2011 4:23pm
Quoting astral77
Your are correct SIR with regard to testing. One thing I have realized during the short time I have been doing forex is like most other skills, there is no substitute for EXPERIENCE. Reading books, articles, forums, seminars, courses they all help (because you are absorbing the fruits of other people`s experience). But at the end of the day nothing replaces what the retail traders call "Screen Time".That is why I highly value the opinion of people like Macman, steve, Bob and your good self to name a few.

Scoobs is in a different league and we...
I will be more than happy if this does work.

Cheers
  • #95
  • Feb 1, 2011 4:32pm Feb 1, 2011 4:32pm
Quoting peterporter
There is an interesting post by ES Trader
Maybe we could incoporate that sort of strategy within the bot

http://www.forexfactory.com/showpost...postcount=2051

Ta
A kind of Recovery for GH. I read ES Trader's post but did not see the significance at the time.

ES trader is another terrific thinker, so we shall give this a whirl. Not immediately because I am not at home tomorrow, but perhaps on Thursday.

In the meantime, perhaps a few of you guys might like to experiment with entry at 90+ and see if it makes a difference.

  • #96
  • Feb 1, 2011 4:36pm Feb 1, 2011 4:36pm
All I am saying is if anyone has some information about it like a study or anything like that I think that would be very helpful. Otherwise, the only thing that I can see to do is to test with stops in place to prevent a large loss.

I am not questioning anyone's knowledge, I just want to know what their opinion is based on. I am still very much up in the air on this method and would appreciate the info if it is out there. Would save me a lot of time programming.
  • #97
  • Feb 1, 2011 5:00pm Feb 1, 2011 5:00pm
Quoting smjones
All I am saying is if anyone has some information about it like a study or anything like that I think that would be very helpful. Otherwise, the only thing that I can see to do is to test with stops in place to prevent a large loss.

I am not questioning anyone's knowledge, I just want to know what their opinion is based on. I am still very much up in the air on this method and would appreciate the info if it is out there. Would save me a lot of time programming.
I sent you a PM
  • #98
  • Feb 1, 2011 5:55pm Feb 1, 2011 5:55pm
Quoting SteveHopwood
A kind of Recovery for GH. I read ES Trader's post but did not see the significance at the time.

ES trader is another terrific thinker, so we shall give this a whirl. Not immediately because I am not at home tomorrow, but perhaps on Thursday.

In the meantime, perhaps a few of you guys might like to experiment with entry at 90+ and see if it makes a difference.

My findings so far are that opening stoch diff at 90 or even 95, and closing at near-0 gives better results.
We must not forget that a stoch diff of 0 only means that both Stochs are just crossing.
But this does not mean that the movement is over: Stochs will eventually continue their way to the opposite side, so the stoch diff should really get negative, but because we are using the absolute value of the stoch diff, we cannot take advantage of this movement continuation: the stoch diff "bounces" on 0 and goes up again, although the movement continues in the same direction, and that is when positive pips are starting to really come in!

There should be a simple way of keeping and using the sign of the stoch diff to our advantage.

Or perhaps simply close trades on a profit target.

We have sort of the same problem with entries : when the stoch diff is beyond, say 90, this only means that the lower stoch is somewhere 0..10, while the higher stoch is 90..100.
But this does not mean the price movement is over and will reverse (that's what we are expecting).
The same issue exists with the Sixths indicator: when price "pushes the walls", the Sixths simply keeps saying "i'm pushing the roof", without being able to tell when it stops pushing... so we are waiting for a retrace below the Gold line to consider this as a significant price reversal.
Technically, the Sixths lines are just a different representation of a stochastic, but behaves exactly the same way: the gold lines are strictly equivalent to stoch levels 16.6 (=1/6) and 83.3 (=5/6).

We should probably wait for each individual stoch to retrace below 90/above 10 once the diff has reached 90, showing that the pairs are starting their way back home.

The other issue with Stochastics is that, when price volatility drops for a significant time compared to the stochastic period (100 bars), the Signal/Noise ratio of the stochastic decreases a lot, meaning that the stoch fluctuations are becoming much more sensitive to small price movements, and that is not good trading conditions: spread cost becomes significant compared to the high/low price levels -- at least when using this method on lower timeframes (M5 or even M1 as i try it).A volatility filter may be useful in this matter (ATR for instance).
I will keep an eye on this to see if i can find anything useful.

We're not there yet...

At the moment, my impression is that we are only playing coin toss...

Scoobs said it all... we are not yet ready to "scalp to death"...

Not enough forward testing yet anyway, so i'm still hopeful!

SQ
  • #99
  • Feb 1, 2011 6:17pm Feb 1, 2011 6:17pm
You said it all here, and you said it well. We are definitely not there yet, not ready to "scalp to death" at all. With this method you'll have lots of little good trades, and one or two horribly bad trades, which will wipe all your profit. At least so far.

I don't personally like the adding on to losing positions that some have advocated; initiating the position at 80%, adding on at 90 and again at 100. The reason is because even though the stoch shows max deviation, prices can actually continue on as they were, and all we're doing is compounding our losing positions (martingaling, if you will) and there is no better chance the latter positions will be profitable than the former.

So what we have here, at best, is a "start" to a system. Numerous things need to change in order to have a good system. And I'm sorry I'm much better at picking out what is wrong than helping to make things right.

Quoting squalou
My findings so far are that opening stoch diff at 90 or even 95, and closing at near-0 gives better results.
We must not forget that a stoch diff of...
  • #100
  • Feb 1, 2011 6:39pm Feb 1, 2011 6:39pm
I think We should prevent going against the trend.

So for example let's say EURUSD is overbought and GBPUSD is oversold and the Stoch Diff is now above 80.

Cool! So we go short EURUSD and long GBPUSD right?

No! We test the trend of both and it seems both pair's trend is UP.
(Which is no surprise as they are positively correlated!)

So what do we do? We just go long with GBPUSD until the Stoch Diff close level is met or until we are in profit.

What do you guys think?
0 traders viewing now