Sixths trading - an EA by macman, Bob and Steve | Page 33 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading the Sixths with Snowballs 68 replies

Gann High-Low System with Laguerre/Fisher/VSA/Sixths and others 124 replies

My Sixths trading journal with TOB EA 50 replies

sixths indicator 1 reply

Steve Trading Journal 3 replies

Attachments: Sixths trading - an EA by macman, Bob and Steve
Exit Attachments

Sixths trading - an EA by macman, Bob and Steve

  • #641
  • Oct 30, 2010 7:29pm Oct 30, 2010 7:29pm
Steve, looks like another system emerged in NB's thread; http://www.forexfactory.com/showthre...46113&page=210 post #3140

Anyway of doing an EA for this? NB likes it
24 hours in a day – 24 bottles of beer in a case – Coincidence? I think not
  • #642
  • Edited 8:08pm Oct 30, 2010 7:39pm | Edited 8:08pm
Quoting scooby-doo
Why dont you use the following for a 7 day high/low being broken for each pair traded instead of using the sixths calculations...

//+------------------------------------------------------------------+
//| expert PivotL function...

I think you are right and maybe this is something to check for each pairs also.

When it moves 0-9 pips from the days opening the probability of it reversing the trend is 75%.

When it moves 10-19 pips from the days opening the probability of it reversing the trend is 55%.

When it moves 20-29 pips from the days opening the probability of it reversing the trend is 40 %.

When it moves 30-39 pips from the days opening the probability of it reversing the trend is 25%.

When it moves 40-49 pips from the days opening the probability of it reversing the trend is 13%.

When it moves 50-59 pips from the days opening the probability of it reversing the trend is 8%.

When it moves 60-69 pips from the days opening the probability of it reversing the trend is 4%.

When it moves 70 pips from the days opening the probability of it reversing the trend is less then 1%.
  • #643
  • Oct 31, 2010 1:33am Oct 31, 2010 1:33am
There is still an issue with thebeast, which i've mentioned previously and that hopefully some of you have also noticed. I do believe it needs corrected, the current gbpchf situation is an example in which it's very obvious, there are other pairs its happening on as well.

The reentrylinepips is being overriden by the 6th indicator recalculating.

What I should have right now on gbpchf is:
- my 1st primary short at 1.5332
- my 2nd short, a reentry at 200 pips (the automatic reentryline pips calculated for this pair) at 1.5532
- my 3rd short, a reentry at another 200 pips at 1.5732, and thebeast should be trying to get breakeven now, as it's set to recover after 3 trades

Instead what I have, due to the 6th indicator recalculating in the middle of my open trades is:
- 1st primary short at 1.5332
- 2nd primary short at 1.5592
- 3rd primary short still pending at 1.5684

Will have to defer to steve or macman regarding this situation to find out if this is something that wasn't intended and needs fixing, or not.
A worse case is in progress on chfjpy, where I have 3 primary buys all about 50 pips apart and in recovery mode now, however these 3 buys should have been 150 pips apart if the 6th repositioning wouldnt have overridden my reentrylinepips of 150.
Hopefully this is clear enough, I can send screenshots if needed.
  • #644
  • Oct 31, 2010 7:27am Oct 31, 2010 7:27am
Guys

I have been very taken with the basket trading idea originally advanced by trade4fun. It has a number of inherent advantages.

macman's dictum that TB should be run on a wide range of pairs is very important as it inherently hedges the aggregate trading position - when one pair is in DD hopefully another is in profit.

The basket idea just takes this one stage further and has the great advantage of requiring just one chart and one EA, so less room for cock-ups. http://cdn.forexfactory.com/images/s...m/big_grin.gif If correctly weighted as well, then the "Law of Large Numbers" for distributions also comes into play (more correctly known as the Central Limit Theorem).

This theorem says (in a handwaving sort of way) that irregularities in a bunch of individual distributions will get evened out in the combined distribution. This kinda side-steps the main problem I have been looking at of trying to optimise the individual pairs. When trading the basket, there is no need to do this.

You can even see by eye from the "T100" chart that the synthetic currency basket mean-reverts much more often than individual pairs. I have tested this formally and am very encouraged by the results.

There is just one small problem with the indy as it stands... It is critically important that the pairs are correctly weighted so the $/pip contribution to the basket is calculated correctly.

I don't believe that the indy is doing this as it stands at the moment. I have been through the code carefully and I think that the calculation for cross-rate pairs is wrong. (I attach a crib-sheet I made to explain the three different pair-types for those who are wondering what I am talking about).

I have an offline system in Matlab that calculates all this (I believe) correctly and I get better results than with the indy.

So.... it may be worth putting some coding effort in here.

To clear up a few previous misconceptions..

(a) yes, an EA runs perfectly on an offline chart - there is no way an EA can even tell it is on an offline chart. E.g. the Renko EA that creates offline M2 charts etc works perfectly with EAs.

(b) no, the basket indy as it stands does not do this as the Window refresh function from the Windows kernel is not called in the same way as the Renko EA - see the MQ4 code of the Renko EA to learn how this should be done properly.

(c) yes, we could turn the basket indy into an EA that would give live-updating offline charts, although the chart would only update when a tick arrived from the pair the indy is hosted on (so EU might be a good choice).

Adapting TB to buy/sell baskets is more tedious than difficult (IMHO) but would be worth doing if the concensus here is that it would give us a significant advantage. I think we should be clear on this before we ask Steve to spend a lot of his valuable time making the changes.

Best regards

Old Dog
Attached Image (click to enlarge)
Click to Enlarge Name: PairTypes.JPG Size: 61 KB
  • #645
  • Oct 31, 2010 7:27am Oct 31, 2010 7:27am
Quoting scooby-doo
Why dont you use the following for a 7 day high/low being broken for each pair traded instead of using the sixths calculations...

//+------------------------------------------------------------------+
//| expert PivotL function...
Here it is. There is little difference between this and simply setting a BarCount of 168.

This version only checks for a new hi/lo at the start of each new D1 candle, so the lines and hence the pending trades are more stable.

It will also have the effect of pushing the intervals between adding new trades wider apart.

I leave it to users to see whether these features are any use or not. Live, I am going with the 'it aint bust' thingy, and will demo this version to see if it makes any difference.

Cheers Scoobs

Attached File(s)
File Type: mq4 The Beast a la Scoobs.mq4   80 KB | 360 downloads
  • #646
  • Oct 31, 2010 8:46am Oct 31, 2010 8:46am
Quoting E2u
My picture is nicer than yours ;-)

Nice chart. So the indicator aggregates all of the pairs you're trading into a single bar chart which you then use the Beast to trade? Aside from simplifying down to one chart I'm having trouble wrapping my head around how this would produce better results.

Thanks.
  • #647
  • Oct 31, 2010 10:02am Oct 31, 2010 10:02am
Old Dog,

thanks for your write up, hopefully steve read upon your message , since he just posted 1 minute after yours. so i guess he could have missed it.

but your explanation is way better then mine.

and i am sure with minds like you or macman you could bring it to a whole other level.
  • #648
  • Oct 31, 2010 10:05am Oct 31, 2010 10:05am
Quoting pipcruiser
I am running the "old" version for 13 trading days on basic pairs and my account is up 90%.
Why? is the Old version any different? I guess you just don' t want to

close all positions to update the Beast, right?

I load Beast ala Scoob on 4 pairs will see what's up..

BTW what lot size you were using?

Thanks.
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #649
  • Edited 12:24pm Oct 31, 2010 11:33am | Edited 12:24pm
Quoting nubchai
Nice chart. So the indicator aggregates all of the pairs you're trading into a single bar chart which you then use the Beast to trade? Aside from simplifying down to one chart I'm having trouble wrapping my head around how this would produce better results.

Thanks.
Its important that the currency pairs you are using in the basket are hedged currency wise.
I use T101 Universal pairs.
GBPUSD EURGBP GBPJPY CADJPY NZDUSD AUDJPY EURJPY EURUSD USDJPY AUDUSD NZDJPY GBPCHF CHFJPY EURCHF
All this pairs are on one offline chart. When you use weighted pairs like this you don't need to think about recovery because no currency pair is left behind. If one pair goes wrong direction an other pair will compensate that, and you open and close the hole basket at the same time.
If the basket was perfect weighted and correlated you would make 0 profit. Now its not the case. Some currency pairs make their own trip out of normal before they comes back to their place in the basket.
With a basket you get a smoother and shorter cycle up and down over the weekly chart.
  • #650
  • Oct 31, 2010 12:02pm Oct 31, 2010 12:02pm
Quoting E2u
Its important that the currency pairs you are using in the basket are hedged currency wise.
I use T101 Universal pairs.
GBPUSD EURGBP GBPJPY CADJPY NZDUSD AUDJPY EURJPY EURUSD USDJPY AUDUSD NZDJPY GBPCHF CHFJPY EURCHF
All this pairs are on one offline chart. When you use weighted pairs like this you don't need to think about recovery because no currency pair is left behind. If one pair goes wrong direction an other pair will compensate that, and you open and close the hole basket at the same time.
If the basket was perfect weighted and correlated...
Do you have a chart showing this shorter and smoother cycling? Pippo
Vincit qui se vincit.
  • #651
  • Oct 31, 2010 12:10pm Oct 31, 2010 12:10pm
Hey

Im running the old version bcs im happy with it

Look at my last few post for more info.

PC

Quoting SkyzerFX
Why? is the Old version any different? I guess you just don' t want to

close all positions to update the Beast, right?

I load Beast ala Scoob on 4 pairs will see what's up..

BTW what lot size you were using?

Thanks.
Less is more...
  • #652
  • Oct 31, 2010 1:48pm Oct 31, 2010 1:48pm
Quoting trade4fun
hi Steve,

thanks for showing intrest.

let me try to explain it again and better.

what is actually needed is that the BEast EA works on an offline chart of T101. so we can go baskettrading.

for example today i opened up buys based on barcount indicator and i am now in 332 pips profit.

please check the picture below as how i did my trades, which i think comfirms the BEast trading logic.

i dont change anything about the entry logic. however for the moment i only use a different exit strategy. since i have an profitbasket Ea. that manages my trades....
Is there any EA to place specified number of currencies order SELL's or BUY's on

the fly? Script perhaps?

Cause it's going to take a while to place 14 trades manually..

Plus let say it's a 3 pips on average spread. 3 *14 = 42 (or more) pips Loss right a way..

Brokers going to be happy. I wish I could make those 42 pips a day guaranteed

Thank you.
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #653
  • Oct 31, 2010 1:55pm Oct 31, 2010 1:55pm
What you can do is play risk aversion T101 basket is basically a risk-on risk-off strategy to capitalize on buying or selling JPY/CHF/USD basically.

If you do the math the BUY basket boils down to buying something like 2 JPY 1 CHF and 3 USD or something like that... can't remember.

So theortically if you chart JPY/USD/CHF against all other commodity currencies in your basket you can get a chart. Use that to trade the BUY or SELL a basket. The idea is that the basket itself represents the market in a better way because your not having to deal with drawdown in case you have a runner that goes against you.. ie in a counter trade scenario you take a SELL basket but GU might keep going up due to stoploss hunting but the rest of the basket retraced enough to cover you.

I think trading a basket is always a good thing because it forces you to think bigger...step back and analyze the entire market that your basket represents.

This only works right now because JPY/USD/CHF are considered safe haven currencies.

I'm not sure if theres a better way to represent the basket of both sides in a single chart because the fundamentals are what I described above. The way to get the best trades out of this using the Beast would be to plot both JPY/USD/CHF and the rest... and take trades when they are on the opposing end of the sixth indicator(or atleast make sure they both aren't overbought/oversold). You wouldn't want to take a SELL basket for example if your commodity currencies are also overbought... they will bring your basket into drawdown and at best you would be breaking even as the JPY/USD/CHF currencies go from overbought to oversold.

I think that covers it.
  • #654
  • Oct 31, 2010 2:01pm Oct 31, 2010 2:01pm
Quoting SkyzerFX
Is there any EA to place specified number of currencies order SELL's or BUY's on

the fly? Script perhaps?

Cause it's going to take a while to place 14 trades manually..

Plus let say it's a 3 pips on average spread. 3 *14 = 42 (or more) pips Loss right a way..

Brokers going to be happy. I wish I could make those 42 pips a day guaranteed

Thank you.
Voila
Attached File(s)
File Type: mq4 UnivTradeHedge.mq4   2 KB | 307 downloads
  • #655
  • Oct 31, 2010 2:05pm Oct 31, 2010 2:05pm
Quoting E2u
Its important that the currency pairs you are using in the basket are hedged currency wise.
I use T101 Universal pairs.
GBPUSD EURGBP GBPJPY CADJPY NZDUSD AUDJPY EURJPY EURUSD USDJPY AUDUSD NZDJPY GBPCHF CHFJPY EURCHF
All this pairs are on one offline chart. When you use weighted pairs like this you don't need to think about recovery because no currency pair is left behind. If one pair goes wrong direction an other pair will compensate that, and you open and close the hole basket at the same time.
If the basket was perfect weighted and correlated...
Thanks E2U that makes sense. Sounds tricky to program on an offline chart. I've used renko bars before so I understand the set-up of the offline chart.
  • #656
  • Oct 31, 2010 2:52pm Oct 31, 2010 2:52pm
Quoting peterporter
Voila
Thanks, Now probably need script to close all open orders at ones

using OrderClose function.. I think I can write that.. no problem..
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #657
  • Edited 3:25pm Oct 31, 2010 3:06pm | Edited 3:25pm
The most important currency pairs in a basket
EURUSD
GBPUSD
————-
USDJPY
EURJPY
————
GBPJPY

The 2 first pairs needs to show same direction as at least 2 of the 3 other.
Then the rest of the basket will follow except USDCHF because its negative correlated to the most important currency pair EURUSD.
Therefore I use a script to sell or buy 13 pairs and I put opposite order on USDCHF. I have only tested it for one week but it worked. I used Steve's Lawgirl's Trend Display when I tested it.
Check so all order close when you use Simple basket Close. If not ,drag it one more time to the chart.
Here is scripts and indicator.
Attached Image
Attached File(s)
File Type: mq4 #BUY13.mq4   2 KB | 303 downloads
File Type: mq4 #SELL13.mq4   2 KB | 302 downloads
File Type: mq4 ##Simple Basket Close v2.mq4   8 KB | 311 downloads
File Type: mq4 Lawgirl's Trend Display v1.00.mq4   11 KB | 332 downloads
  • #658
  • Oct 31, 2010 3:30pm Oct 31, 2010 3:30pm
I just construct Close_All open orders at ones SCRIPT

Cannot test it right now (no errors in MetaEditor),

please check and let me know, if

you think there is a problem.. Thanks
Attached File(s)
File Type: mq4 Close_all_T101.mq4   2 KB | 311 downloads
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #659
  • Oct 31, 2010 3:33pm Oct 31, 2010 3:33pm
By the way there is already newer version of LawGirl Trend display

here

P.S.

E2u Simple basket close, script to complected.. I think

all you need is just close all open orders at ones, and that's it..

That's what I've tried too do in my small script there..

P.S.S

Besides:

EURCHF and EURGBP it's not the real pairs, it's just a MIX... I want to exclude those for sure

Same goes for CHFJPY
Attached File(s)
File Type: mq4 Lawgirl's Trend Display v1.02.mq4   13 KB | 395 downloads
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #660
  • Oct 31, 2010 3:41pm Oct 31, 2010 3:41pm
[quote=SkyzerFX;4135504]By the way there is already newer version of LawGirl Trend display

Thanks!! I have missed that.



Its something strange in the Simple Basket Close script. It always leave 1 or 2 orders so I have to drag and drop it again
2 traders viewing now