Is This My Holy Grail EA :D | Page 38 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

The HOLY GRAIL 130 replies

Developing the "Holy Grail" Trading System 4 replies

Holy grail trading system wanted, or nearest thing to it! 43 replies

6 Steps To Finding Your Personal Holy Grail 6 replies

The Search for the Holy Grail 1 reply

Attachments: Is This My Holy Grail EA :D
Exit Attachments

Is This My Holy Grail EA :D

  • #741
  • Aug 29, 2008 4:52am Aug 29, 2008 4:52am
Quoting steveshutts
Bob,

You didn't mention how you produced your equity graphs.
Also, you can run 0.01 trades with a $3000 account, so i was wondering if you had a $10000, coudl you run it with 0.03 i wonder?

Also, how do you start your demo accounts with specified balances, different from the $5000 you usually get?

Steve

Edit: Sorry, just seen your post above. Still would like to know the answer to the last question though, thanks.
Steve,

On MT4 platform, click file>Open an account.

In the popup, on the lower right corner beside "Deposit" on the drop-down menu it shows several default account balances to start with. Instead of choosing one, just type in the amount you want to start with.

Hope that answers your question.

PD
  • #742
  • Aug 29, 2008 10:48am Aug 29, 2008 10:48am
Quoting nanningbob
That is in the EA but I have never used or tested it.

Hi Bob:

May you tell me in the expert the name of the option where I can take the profit and don't star trading again.

Thank you Bob

Ricardo
  • #743
  • Edited 12:41pm Aug 29, 2008 12:40pm | Edited 12:41pm
Quoting nanningbob
Properly killing death trades does not kill the account or cut its profits. Some may set you back a couple of days but usually make a profit for the week. I just showed how I handled the GBP over the weekend and got out for a small loss but the other currencies profits allowed the account to make a profit for the day. We call it a death trade but what we are doing is killing it before it actually becomes one. Death trades takes days or even weeks to wipe out an account. You catch it early and take a loss like you would with any other trading system. They just happen so infrequently that they dont pose too much of a problem. This EA has a lot of play room.
Hi Bob,

I've just started reading this thread and its very educational to say the least for anyone who do not really understand the Martingale system. Thanks for all your input.

I find the death trades very interesting which begs the question for me. Is there a way to have an exit strategy when such conditions occur as you described and also had charts posted with the description of the death trade?

Looking forward to your answer.

Regards,
CT
  • #744
  • Aug 29, 2008 5:20pm Aug 29, 2008 5:20pm
Quoting PistolDave
Again, here is an update to my test of the system.

This is a variation of nanningbob's system that I am trying.

I am using the EAs but, manually opening and closing trades in a little different way than what he has shown to work just to see what happens.

He has a proven, profitable way of trading. I have tested his way also and it has worked well for me just like it is working for him and others.

I will continue to test the strategy using some different variables just to see if I can discover the most effective way to use the system.

Note: the drawdowns I have had with this variation are much higher and the results are much more erratic than I had when using the EAs "properly" as nanningbob has been doing.

Here are the results so far, using the three demo accounts:


Also see the previous posts for exact details on the test:
page 44, post #652
page 46, post #681
page 48, post #707
Latest Updates to test:

Date Acct#1 Acct#2 Acct#3
Aug 20: -$9.75 -$34.99 -$51.74
Aug 21: -$82.54 -$433.23 -$25.74
Aug 22: -600.21 -271.11 -$512.32
Aug 25: +$1.86 +$80.43 -$6.50
Aug 26: +$233.81-$93.48 +$165.31

Aug 27: +$425.71 ...... +$474.78
Aug 28: -$50.82 ...... +$11.66
Aug 29: -$60.38 ...... -$51.95

The missing data (......) is from the Acct#2 which I decided to close after 5 trading days. It was not from a blown account, but rather that it was the slowest gains of all the tests and this is the one that I had been opening and closing twice daily. It shows that that method limits the losses the same as closing all positions once a day, but also keeps the EA from being able to stay open long enough to take advantage of moves that occur that take more than several hours to develop.

Good information. I will continue to test the other two accounts for the next few trading days. So far they seem to be about similar in results. I will explain the variations in a few days. Right now I am leaning toward leaving them open more than one day without closing them to see how they perform. It seems that if you are going to open/close on a regular schedule, without using any discretion on closing the individual pairs manually, then to keep the EAs running for as long as possible "hands off" is the most profitable way.

That will probably be my next test.

Thanks again to nanningbob for working with me on this test - and for introducing this strategy to us in the first place.

Good holiday weekend to you,
PD
  • #745
  • Aug 30, 2008 5:04am Aug 30, 2008 5:04am
Quoting cashflowtrader
Hi Bob,

I've just started reading this thread and its very educational to say the least for anyone who do not really understand the Martingale system. Thanks for all your input.

I find the death trades very interesting which begs the question for me. Is there a way to have an exit strategy when such conditions occur as you described and also had charts posted with the description of the death trade?

Looking forward to your answer.

Regards,
CT
Yes they are in the thread but I am in Hong Kong and dont have time to get answer now. I will return in a week.
  • #746
  • Aug 30, 2008 5:09am Aug 30, 2008 5:09am
Quoting fotoposter1
Hi Bob:

May you tell me in the expert the name of the option where I can take the profit and don't star trading again.

Thank you Bob

Ricardo
When you drag the EA on the screen it will give you options for trading. Click on tabs, scroll down to 'my money target' change it from false to true. The next line says 'my_money_target' put the amount in and try it. I have never tested this option so demo test it and see if it does what you are looking for. You will also see dozens of other choices you can play with. The sto settings however are all disconnected.
  • #747
  • Aug 30, 2008 10:47am Aug 30, 2008 10:47am
Hi Bob
Public results of my demo.
The first, running from three weeks, as you know, not allow me to save the file and I do not know why.
Then write the date here:
Departure from 9000.00 dollars, the first operation of 23.03 hours 10-08-2008, last Friday 29-08-2008 hours 18.58
Balance 11471.62 - equivalent 11479.28 - 7.76 float
Work on Couples: EUR-USD/USD-CHF/GBP-USD/USD-JPY/GBP-JPY
I can not see the maxDD.
The second demo "exstreme" (see my post 695) I had to leave because my line does not hold many cross .. (46) (I live in the countryside and the connection is defective)
I set up two demo that you have done for me and the results they see below. They work both on 26 couples with spre <= 12
One question:
Why not ever reply to my personal letters?
thank you
Romano
Attached File(s)
File Type: zip settimana 25-29.zip   25 KB | 315 downloads
  • #748
  • Edited 3:41pm Aug 30, 2008 3:18pm | Edited 3:41pm
I was intruiged by Bob's statement that it should not cost very much to cut a potential death trade. I tried to quantify some of the potential losses and have to say - either I am doing something wrong or it does not look so encouraging.

I attach files with

(a) An estimate of the maximum loss you would have at each level if you are on the wrong side of the trend and it keeps going against you (e.g. $10 per pip as in EURUSD). This is based on 50/40 at multiplication of lot size by 1.8.

(b) Bob's 7-yr stats on how much pairs fluctuate on one given day, sorted by number of events with days of >250 pips.

I see very large costs for getting out of trades once you get beyond about 250 pips. Because the multiplication factor almost doubles the lot size with each step, the losses for 0.01 beginning lot size stay under $1,000 below about -200 pips against you, and then explode to over $8,000 as you approach -350 pips.

How rare is it that you will have to cut losses?

The 7-yr stats are saying that 250 pip movements are fairly frequent. Frequency only peters off after 300-350 pips for the low-moving pairs in the upper half (sorted by number of 250 pip days). A considerable number of pairs are having many days with even higher fluctuations!

A quick estimate may show that events exceeding -300 pips occur on average 2-3 times a year for the few low ranging pairs, 6-7 times a year for mid-ranging, and much more often for high-ranging pairs (particularly including the Yen).

So it appears that it all depends on the question how much you can reduce the probability of a >250 pip event per day, by closing all trades every day. I would think there is a real drop in probability by doing so - but it is extremely hard to measure. How often will you be on the wrong side of the trend when these landslides happen, even if you closed the trades the day before?

So this exercise leaves me a little discouraged. It looks like that this daily landslide band of 250-350 pips is exactly the zone where things are balanced - and a shift in tiny probabilities of being on the wrong side will have huge consequences.

If you are a good manual trader like Bob you may be able to cut the cost of killing a potential death trade considerably - I am not sure if I can.

It is very hard to test these things. Just forward testing and having faith that things will be OK starts looking more and more like gambling to me.

I am playing devil's advocate here and would love to be proven wrong. I would appreciate your feedback on this perspective - perhaps I made an error or overlooked an important aspect.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: MM_Max Loss.JPG Size: 211 KB
Click to Enlarge Name: 7-yr Stats.JPG Size: 250 KB
  • #749
  • Aug 30, 2008 4:54pm Aug 30, 2008 4:54pm
Quoting bubo
I was intruiged by Bob's statement that it should not cost very much to cut a potential death trade. I tried to quantify some of the potential losses and have to say - either I am doing something wrong or it does not look so encouraging.

I attach files with

(a) An estimate of the maximum loss you would have at each level if you are on the wrong side of the trend and it keeps going against you (e.g. $10 per pip as in EURUSD). This is based on 50/40 at multiplication of lot size by 1.8.

(b) Bob's 7-yr stats on how much pairs fluctuate on one given day, sorted by number of events with days of >250 pips.

I see very large costs for getting out of trades once you get beyond about 250 pips. Because the multiplication factor almost doubles the lot size with each step, the losses for 0.01 beginning lot size stay under $1,000 below about -200 pips against you, and then explode to over $8,000 as you approach -350 pips.

How rare is it that you will have to cut losses?

The 7-yr stats are saying that 250 pip movements are fairly frequent. Frequency only peters off after 300-350 pips for the low-moving pairs in the upper half (sorted by number of 250 pip days). A considerable number of pairs are having many days with even higher fluctuations!

A quick estimate may show that events exceeding -300 pips occur on average 2-3 times a year for the few low ranging pairs, 6-7 times a year for mid-ranging, and much more often for high-ranging pairs (particularly including the Yen).

So it appears that it all depends on the question how much you can reduce the probability of a >250 pip event per day, by closing all trades every day. I would think there is a real drop in probability by doing so - but it is extremely hard to measure. How often will you be on the wrong side of the trend when these landslides happen, even if you closed the trades the day before?

So this exercise leaves me a little discouraged. It looks like that this daily landslide band of 250-350 pips is exactly the zone where things are balanced - and a shift in tiny probabilities of being on the wrong side will have huge consequences.

If you are a good manual trader like Bob you may be able to cut the cost of killing a potential death trade considerably - I am not sure if I can.

It is very hard to test these things. Just forward testing and having faith that things will be OK starts looking more and more like gambling to me.

I am playing devil's advocate here and would love to be proven wrong. I would appreciate your feedback on this perspective - perhaps I made an error or overlooked an important aspect.
Exiting on a retrace stops the big loss. You are assuming in your figures the max on a loss. The skill is exiting on a retracement before before you go outside the EAs range. Currencies always retrace, unless someone drops a N-bomb someplace. Also, I use pennies and not dimes, or dollars. Makes all the figures change and changes the risk factor also.
  • #750
  • Aug 30, 2008 5:11pm Aug 30, 2008 5:11pm
I did a live retracement while the trade was happening of how to handle a potential death trade. It was the big GBP/USD move at the time. It started on p 47 post 694 and ended on P48 post 706. That is how you avoid the max losses mentioned earlier.
  • #751
  • Aug 30, 2008 5:14pm Aug 30, 2008 5:14pm
Quoting cashflowtrader
Hi Bob,

I've just started reading this thread and its very educational to say the least for anyone who do not really understand the Martingale system. Thanks for all your input.

I find the death trades very interesting which begs the question for me. Is there a way to have an exit strategy when such conditions occur as you described and also had charts posted with the description of the death trade?

Looking forward to your answer.

Regards,
CT
See post
#750 for instructions
  • #752
  • Sep 1, 2008 4:39am Sep 1, 2008 4:39am
Bob,

Seeing as we are trading all currencies, some of these will be highly correlated. Have you ever experienced multiple death trades occuring during your testing. Say, for instance, we get on the wrong side of the Gbp/Yen and the Eur/Yen and they both really move. Would multiple death trades wipe the account?

Steve
  • #753
  • Sep 1, 2008 5:13am Sep 1, 2008 5:13am
Quoting steveshutts
Bob,

Seeing as we are trading all currencies, some of these will be highly correlated. Have you ever experienced multiple death trades occuring during your testing. Say, for instance, we get on the wrong side of the Gbp/Yen and the Eur/Yen and they both really move. Would multiple death trades wipe the account?

Steve
Even thought that is true, how a currency retraces though varies. I have seen gbp/jpy dive retrace and then go on the right side of the trade and the eur/jpy dive and hold steady. Their general direction maybe the same but their moment by moment will vary significantly. Again, it has to do with retracements. If you read this post I have only dealt with two possible death trades in the last several weeks. The gbp/usd and the nzd/usd. I wrote about both and they both had happy endings. In my testing I have not faced a multiple death trade scenario from similiar pairs because all currencies will retrace. That comes from 9 months of testing now. The USD made a major move in AUG. improving on every single currency but only the GBP/USD was the major test. The NZD/USD was being tested on a 50/20 and went out of range on that but not on my 50/40 or 100/40. Good question.

I visited the visa office today and things went well. Now I wait for their answer in a couple of days. Gonna take some time off and enjoy the sights.
  • #754
  • Sep 1, 2008 7:49pm Sep 1, 2008 7:49pm
Quoting PistolDave
Latest Updates to test:



The missing data (......) is from the Acct#2 which I decided to close after 5 trading days. It was not from a blown account, but rather that it was the slowest gains of all the tests and this is the one that I had been opening and closing twice daily. It shows that that method limits the losses the same as closing all positions once a day, but also keeps the EA from being able to stay open long enough to take advantage of moves that occur that take more than several hours to develop.

Good information. I will continue to test the other two accounts for the next few trading days. So far they seem to be about similar in results. I will explain the variations in a few days. Right now I am leaning toward leaving them open more than one day without closing them to see how they perform. It seems that if you are going to open/close on a regular schedule, without using any discretion on closing the individual pairs manually, then to keep the EAs running for as long as possible "hands off" is the most profitable way.

That will probably be my next test.

Thanks again to nanningbob for working with me on this test - and for introducing this strategy to us in the first place.

Good holiday weekend to you,
PD
Update to the testing.

Added data from Sept. 01 2008:

Date Acct#1 Acct#2 Acct#3
Aug 20: -$9.75 -$34.99 -$51.74
Aug 21: -$82.54 -$433.23 -$25.74
Aug 22: -600.21 -271.11 -$512.32
Aug 25: +$1.86 +$80.43 -$6.50
Aug 26: +$233.81-$93.48 +$165.31

Aug 27: +$425.71 ...... +$474.78
Aug 28: -$50.82 ...... +$11.66
Aug 29: -$60.38 ...... -$51.95
Sept 01: +$201.18 ...... +$149.20

Things are running smooth.
PD
  • #755
  • Sep 2, 2008 1:39am Sep 2, 2008 1:39am
Yes, my babies are running smooth as well! I am running a 40/100, and a 40/50, which i monitor a few times a day to see the profit and when it looks good i take it.
I have also just started a new 40/100, which i will leave to run for a couple of days before restarting to see whether this reaps more profit, although i will close if i see huge profits. I started this last night, beginning of Asian and looked at it this morning to see $179 profit on a $3000 account, 5.9% for sleeping, that will do, LOL, so i have closed that for the rest of the day. The bummer was, my other 40/100 had not traded for some reason, i think the app had frozen, which was annoying as that would have really boosted my already very profitable account. Still, these things happen!
I switched my 40/50 off at $80 profit and the other 40/100 has just started running.
I am desperately waiting for some death trades to see how i cope as this EA seems to generate profits every day at some time and it just seems easy money at the moment, but i know that could easily change. Its all about how a death trade is dealt with, so bring it on!

Edit: I think it is definitely an advantage to be able to constantly monitor the profit level during the European and NY Sessions as these can really fluctuate. At the moment, i can do this. I am therefore going to run the other EA and only try to look at it a couple of times a day. I don't really even look at the charts much, i just look at open and closed trades and see which levels price has got to, ie if they are in danger or not.


Steve
  • #756
  • Sep 2, 2008 11:13am Sep 2, 2008 11:13am
Well, after starting my 40/100 late this morning, and missing all the large profits that my other 40/100 had reaped during the Asian session, i ended up being on the wrong side of the Aussie currencies all day. I was long on most of the Aussies when it was crashing. I went to about 6 levels deep on a couple of them so was starting to monitor these for retracements.
Just came back after a long lunch and everything was in a healthy profit after experiencing some nice retracements. Just shows how you need to wait some time with these drawdowns.
Just closed with $106 profit for the day, 3% so very nice.

Steve
  • #757
  • Sep 2, 2008 11:17am Sep 2, 2008 11:17am
Quoting steveshutts
Well, after starting my 40/100 late this morning, and missing all the large profits that my other 40/100 had reaped during the Asian session, i ended up being on the wrong side of the Aussie currencies all day. I was long on most of the Aussies when it was crashing. I went to about 6 levels deep on a couple of them so was starting to monitor these for retracements.
Just came back after a long lunch and everything was in a healthy profit after experiencing some nice retracements. Just shows how you need to wait some time with these drawdowns.
Just closed with $106 profit for the day, 3% so very nice.

Steve
Hi Steve:

Are you using mini lots .01 in Interbankfx 200:1 or you are .1 ?

Thank you
Ricardo
  • #758
  • Sep 2, 2008 12:37pm Sep 2, 2008 12:37pm
Yes, i am using 0.01 lots on all the ccy's, that is plenty! Its reaping rewards at the moment, everyday is profitable but i have had no problems yet. The most extreme so far is about 6 levels so waiting for the death trade so i can experience it. My account is up $411 in 6 days of trading, so about 13%. Not bad for trading 0.01 lots.
  • #759
  • Sep 3, 2008 6:13am Sep 3, 2008 6:13am
Quoting PistolDave
Update to the testing.

Added data from Sept. 01 2008:



Things are running smooth.
PD
Added data from Sept 02 2008 & 10 day trading summary:


Date Acct#1 Acct#2 Acct#3
Aug 20: -$9.75 -$34.99 -$51.74
Aug 21: -$82.54 -$433.23 -$25.74
Aug 22: -600.21 -271.11 -$512.32
Aug 25: +$1.86 +$80.43 -$6.50
Aug 26: +$233.81-$93.48 +$165.31

Aug 27: +$425.71 ...... +$474.78
Aug 28: -$50.82 ...... +$11.66
Aug 29: -$60.38 ...... -$51.95
Sept 01: +$201.18 ...... +$149.20
Sept 02: +$48.61 ...... +$27.77


For the complete details of this test please see page 44, post 652.

Test results, last 10 consecutive trading days:

Account #1: +$335.04
Account #2: Closed before end of test because of lagging performance.
Account #3: +$530.29

(Note: The reason these numbers don't equal the sum of the days trading results is because I forgot to post some of the profit from the trading day Aug 21. My mistake.)

The more important fact is that there was NO floating drawdown on the account at the end of any day to distort the results, and that there was a relatively steady increase in the balance.

These are graphs the latest from the end of the trading day yesterday - and all open positions had been closed out and the balance was squared with the equity at the end of each day - no positions were left open more than 24 hours.)

Test results, over longer period:

Test results, Account #1, 17 consecutive trading days p/l: +$1904.94
Test results, Account #2, 13 consecutive trading days p/l: +$922.82

Started with $3000 in each account.

Summary:

Excellent results...

Very similar results in the two accounts over the 10 day and longer period. Except for one day where the drawdown was $500-$600 (Aug 22) there was no significant setback on either, and the recovery was rapid. The general trend in the equity curve (except for some flat periods and the drawdown) has been upward. The profit factor is 1.58 on 1018 trades on Acct #1 and 1.47 on 509 trades on Acct #2 which seems very good.

From what I have seen so far, I consider this to be a conservative way to trade these EAs and this strategy. In fact, while increasing the safety, resetting all the positions every day would seem to unnecessarily constrain the ability of the EAs to produce profit. I think the more reasonable way to trade them would be to let them continue to work their way along with the trend and only monitor the positions once or twice daily to close out the individual positions that have reached a predetermined drawdown amount and only close those positions as needed.

I will start testing that strategy right away. I think that is the method Bob has been using with good results.

I can also say I am leaning toward using the strategy in Acct #2 (using the 100x40 on all pairs, high and low volatility) which is Bob's method, since it seems to produce slightly higher profits and reduces some of the drawdown.

The graphs shown below are from Acct #1, all 17 days since it was opened (1018 trades), and from Account #2, all 13 days since it was opened (509 trades).

I will continue to post my results of the testing.

** Thanks again Bob for developing this amazing trading strategy. **

Good luck,
PD
Attached Image(s) (click to enlarge)
Click to Enlarge Name: DetailedStatement18 - Acct1 - 17day.gif Size: 5 KB
Click to Enlarge Name: DetailedStatement8 - Acct2 - 12day.gif Size: 5 KB
  • #760
  • Sep 3, 2008 7:44am Sep 3, 2008 7:44am
Quoting steveshutts
Yes, i am using 0.01 lots on all the ccy's, that is plenty! Its reaping rewards at the moment, everyday is profitable but i have had no problems yet. The most extreme so far is about 6 levels so waiting for the death trade so i can experience it. My account is up $411 in 6 days of trading, so about 13%. Not bad for trading 0.01 lots.
Hi Steve:

May you add your statement, because I dont have that profit, maybe I am wrong with the size of the lots.

Another question; are you enter just with the direccion of the trend with 4 h, because not always are the pairs over 80 or under 20 level. How many ccy's are you trade every day?

Thanks
Ricardo
1 trader viewing now