Is This My Holy Grail EA :D | Page 13 | 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

  • #241
  • Feb 25, 2008 6:22am Feb 25, 2008 6:22am
Quoting frito
yeap definitely not one fit for news.

Did not all the death trades happen during the news or due to holding the position through the weekend?
I cant say absolutely for sure but whenever I studied the death trade on the back tests they always ran for several days or weeks meaning there was always time to get out of them. When you see price moving up at a nice angle and then a sudden drop look at when the first trade entered and see how long it took to wipe out the account. Most of the time it took a long period of time. I then began developing a strategy of using the Martingale system only for day trading and not holding positions for more than one day. I have found it has worked quite well. I think it is workable strategy this way and you really do not have to worry about the death trade if your funds to trade ratios are right. Too many guys used 1 or 2,000 dollars and played 1.00 per pip and just got killed. They assumed it was the trade system but their ratios werent right and they got wiped out.
  • #242
  • Feb 25, 2008 6:32am Feb 25, 2008 6:32am
Quoting Numeral
Hi Bob,
Good point of view,
I have some times better results with sto and other times with MA.

What do u think about using a weekly pivot and a very very smal lot to handle - 500 pips ?
I have not thought of that. I was thinking for the GBP/JPY of using the daily trend and being able to go 5-600 deep (reenter every 50 points TP every 75-100 points) The real problem I have with GBP/JPY is the 9 pip spread the IBFX uses. I wish they would lower their pip spread on that pair along with he AUD/JPY and NZD/AUD. The NZD/AUD has always tested the best of all the currencies but the 11 point spread scares me.
  • #243
  • Feb 26, 2008 6:41am Feb 26, 2008 6:41am
Quoting nanningbob
I have not thought of that. I was thinking for the GBP/JPY of using the daily trend and being able to go 5-600 deep (reenter every 50 points TP every 75-100 points) The real problem I have with GBP/JPY is the 9 pip spread the IBFX uses. I wish they would lower their pip spread on that pair along with he AUD/JPY and NZD/AUD. The NZD/AUD has always tested the best of all the currencies but the 11 point spread scares me.
Bob, have you a list of what is the better TP and PipStep of each pair? Maybe where we can get the info to create it.

For long time trade (different from your strategy, trading only for day)
i find for NZD/JPY tp 30, step 20
  • #244
  • Feb 26, 2008 5:00pm Feb 26, 2008 5:00pm
Quoting Numeral
Bob, have you a list of what is the better TP and PipStep of each pair? Maybe where we can get the info to create it.

For long time trade (different from your strategy, trading only for day)
i find for NZD/JPY tp 30, step 20
That is the process I am working on now. I am trying 25 tp 15 step with the slower currencies and 50 TP 20 step on the more volatile currencies. I set up a 20 TP step 7 and I have tried a 15 tp step 5. The 15 tp 5 step worked great most days but can get into trouble too easily. Surprise news stories can kill that one. One thing I like to do is find a combo of 2-4 currencies that compliment each other. Like the Eur/Usd, Eur/chf, eur/jpy and Nzd/USD. That combo seems to work. They all move for different reasons and different times during the day. I have the 25 tp 15 step on 3 currencies and the 50 tp 20 step on the eur/jpy. Using pennies it did over 15.00 profit in one day. I am also demo testing .10 trades using 3000 on some combinations also. I think the strategy is sound it is now getting the right combos with the right currencies. 30 tp step 20 gives you protection over 250 pips and that should work well with that currency pair.
  • #245
  • Feb 27, 2008 4:41am Feb 27, 2008 4:41am
Quoting nanningbob
That is the process I am working on now. I am trying 25 tp 15 step with the slower currencies and 50 TP 20 step on the more volatile currencies. I set up a 20 TP step 7 and I have tried a 15 tp step 5. The 15 tp 5 step worked great most days but can get into trouble too easily. Surprise news stories can kill that one. One thing I like to do is find a combo of 2-4 currencies that compliment each other. Like the Eur/Usd, Eur/chf, eur/jpy and Nzd/USD. That combo seems to work. They all move for different reasons and different times during the day. I have the 25 tp 15 step on 3 currencies and the 50 tp 20 step on the eur/jpy. Using pennies it did over 15.00 profit in one day. I am also demo testing .10 trades using 3000 on some combinations also. I think the strategy is sound it is now getting the right combos with the right currencies. 30 tp step 20 gives you protection over 250 pips and that should work well with that currency pair.
One idea is use always correlation pairs, buying one and selling other. looking in a retrace for both.. if one retrace the other retrace too, i they go over the steps, they r protected by correlation.

- Protection over 250, so you r using 12 levels of step? with 1.6x factor, sl 20 pips, go to 2.81 lots. maybe it can handle a maximum of 5 pairs together? what do u think?
  • #246
  • Feb 27, 2008 5:04pm Feb 27, 2008 5:04pm
Quoting Numeral
One idea is use always correlation pairs, buying one and selling other. looking in a retrace for both.. if one retrace the other retrace too, i they go over the steps, they r protected by correlation.

- Protection over 250, so you r using 12 levels of step? with 1.6x factor, sl 20 pips, go to 2.81 lots. maybe it can handle a maximum of 5 pairs together? what do u think?
Yes when I say protection to 250 I am assuming you can cover comfortably 12 levels. I do not let it enter any more than 12 levels because then you have to worry about wiping out your account. I can handle a 12 level loss both monetarily and psychologically. 5 pairs, hmmmmmm, I am live testing 4 pairs and on my demos I am using 6. I was happy when the tp25 step 15 survived the 200+ move on the eur/usd. It retraced enogh to close a trade and then it reentered in the same direction as the move. Every 25 points after that was a profit. That gave me a lot of confidence in that figure for that pair. You wont get manydays like that. It also survived a 135+ almost straight drop on the USD/CAD pair. That one turned a nice profit when it retraced. I will try to give a 10 day report on my live and demo trades this weekend. The others can try them on their demos and compare.
  • #247
  • Feb 27, 2008 5:41pm Feb 27, 2008 5:41pm
Quoting Numeral
One idea is use always correlation pairs, buying one and selling other. looking in a retrace for both.. if one retrace the other retrace too, i they go over the steps, they r protected by correlation.

- Protection over 250, so you r using 12 levels of step? with 1.6x factor, sl 20 pips, go to 2.81 lots. maybe it can handle a maximum of 5 pairs together? what do u think?
If you use two currencies that move in correlation I wouldnt use more than two together. That means both can go into a death trade together. When I think of correlation I am thinking of currencies that dont move together. Like the USD/CAD generally moves during the US trade period while the EUR/CHF moves strongest during the Euro session. I like EUR/JPY because it moves in all sessions. So pairing the NZD/USD or aud/usd with a EUR or CHF pair will have them moving at different times during the trading day. You can have multiple moves during the day but not at the same time helping to protect your account. That is the idea I am working with.
  • #248
  • Feb 28, 2008 5:53am Feb 28, 2008 5:53am
Ok Bob,

another thing.. what was the change u make at Power Stoch?
Did you change anything in the entrys ?

Regards,
  • #249
  • Feb 28, 2008 6:36am Feb 28, 2008 6:36am
Quoting Numeral
Ok Bob,

another thing.. what was the change u make at Power Stoch?
Did you change anything in the entrys ?

Regards,
I am using mostly MA systems following the trend on the 4 hour charts. I have been doing one demo using a sto setting but I keep running into bad trades now and then. MA cross follows trends better than overbought/oversold of the sto. I think the longer time period stos like 21.5.3 work better than the quicker ones like a 14.3.3 I have also had better success on the 30M sto rather than the 1H. But like I said most of my work is using MA cross on the 4H chart. I also keep the trades running so as soon as I take profit it immediately reenters in the trend direction of the 4H chart.
  • #250
  • Feb 28, 2008 11:14am Feb 28, 2008 11:14am
Quoting nanningbob
I am using mostly MA systems following the trend on the 4 hour charts. I have been doing one demo using a sto setting but I keep running into bad trades now and then. MA cross follows trends better than overbought/oversold of the sto. I think the longer time period stos like 21.5.3 work better than the quicker ones like a 14.3.3 I have also had better success on the 30M sto rather than the 1H. But like I said most of my work is using MA cross on the 4H chart. I also keep the trades running so as soon as I take profit it immediately reenters in the trend direction of the 4H chart.
So you use 4H trend, this give you a long trend direction, how u avoid death trade at wrong entrys in this long trends? Manualy close or use any kind of SL?

Using only one MA price cross or MA´s crosses?

Thanks! Regards,
  • #251
  • Feb 28, 2008 8:38pm Feb 28, 2008 8:38pm
Bob,

I was thinking... did u use the 10point3 v.0.02 ? I think he works in that way... always opening trades in the direction of trend.. but uses MA of MACD.

Or was other?

Regards
  • #252
  • Feb 29, 2008 5:13am Feb 29, 2008 5:13am
Quoting Numeral
So you use 4H trend, this give you a long trend direction, how u avoid death trade at wrong entrys in this long trends? Manualy close or use any kind of SL?

Using only one MA price cross or MA´s crosses?

Thanks! Regards,
I like 4H trend the best. It gives me the most consistent market trend. Use a 1MA of the high/low crossing the MA 10 shift 1 open. If you have your period separators on, the period from white line to white line is 1 week on a 4H chart. You will see it will only cross 1-3 times a week as an average. I also balance it with a cross of the 1MA and 3MA on the 5M. My testing shows I get far fewer death trades because price is always following the trend and when there is a trend switch that is where you have the battle between the bears and the bulls so you get the quick close out and you are back with the trend again. Today was a perfect example, all my live accounts and demos that followed the 4H trend did great and those that were on Sto got wiped by the big moves in the JPY NZD and AUD currency pairs. My live accounts had their best day of the week today and I have doubled my live account for the week. I took my wife out for dinner tonight to celebrate. I also have used the 10.3 and found the same problem with it as I did with the sto indicator. The Sto gives change of direction too quickly and you end up on the wrong side of the trend too often. The MACD takes too long for the trend change and you end up on the wrong side of the trend too. A 4H trend turn around can be as short as 12 hours and sometimes be longer than 48 hours so there is plenty of time to recognize it and be ready. Actually I have something better. I will share in my next post.
  • #253
  • Edited 5:36am Feb 29, 2008 5:32am | Edited 5:36am
I am attaching what I consider the most important information needed for Martingale trading. Fundamental Analysis is basically useless for Martingale trading and technical analysis is only useful for following a trend. It makes no difference what technical indicator you use, none have proven totally reliable in trading or there wouldn't be thousands of them out there. Therefore a new philosophy or attitude of trading must be used to make a martingale reliable and usable. I posted earlier information about 16 currencies and their results for the last 7 years. How many pips did they move in one day and how often did they go over an extreme. Attached is the information of how many times they went over 100 pips, 150, 200, 300, 400, etc. Instead of testing and retesting certain currencies with certain formulas. You can look at the information and develop a formula to fit a currency or certain currencies. For instance; the eur/gbp had 7, chf/jpy had 22, eur/chf had 32 days in 7 years of going over 150 pips in one day. That means if you set up a reentry every 15 pips with these pairs, the chances of going into a death trade is almost 0. If you reenter every 10 points you can probably surmise of very little chance of facing the death trade over the year, and that is assuming it did 150 points without retracing, a very rare event in currency trading. So here is the info and use it to develop your strategy. I think you will be surprised that some currencies do fit together and could use the same formula successfully. I color coded some of the groupings for an easier read. Happy trading.

Bob
Attached File(s)
File Type: xls 7 year currency facts.xls   16 KB | 742 downloads
  • #254
  • Feb 29, 2008 6:11am Feb 29, 2008 6:11am
Quoting Numeral
Bob, have you a list of what is the better TP and PipStep of each pair? Maybe where we can get the info to create it.

For long time trade (different from your strategy, trading only for day)
i find for NZD/JPY tp 30, step 20
A step 20 going 12 levels puts you at 240 points+. In the last 7 years NZD/JPY went over 250 pips in one day only 36 times. Thats an average of 5 times a year. Add to that, the possibility that while it was moving over 250 pips, that somewhere it would retrace the 30+ you need to take profit, I would say your chances of a death trade on this pair would be near 0. I think you could safely trade that pair with those figures 24/7. If a death trade were to develop, it would take over 24 hours to develop and you would have plenty of time to recognize it and plan an exit strategy. Good luck. Thanks for the figures, I may try that one on a demo. If you look at my Excell chart I posted previously, you will see that the USD/CAD, EUR/USD, AUD/JPY, and the USD/JPY post approximately the same numbers so those currencies would have similiar odds and probably would work also.
  • #255
  • Feb 29, 2008 6:25am Feb 29, 2008 6:25am
Quoting nanningbob
I am attaching what I consider the most important information needed for Martingale trading. Fundamental Analysis is basically useless for Martingale trading and technical analysis is only useful for following a trend. It makes no difference what technical indicator you use, none have proven totally reliable in trading or there wouldn't be thousands of them out there. Therefore a new philosophy or attitude of trading must be used to make a martingale reliable and usable. I posted earlier information about 16 currencies and their results for the last 7 years. How many pips did they move in one day and how often did they go over an extreme. Attached is the information of how many times they went over 100 pips, 150, 200, 300, 400, etc. Instead of testing and retesting certain currencies with certain formulas. You can look at the information and develop a formula to fit a currency or certain currencies. For instance; the eur/gbp had 7, chf/jpy had 22, eur/chf had 32 days in 7 years of going over 150 pips in one day. That means if you set up a reentry every 15 pips with these pairs, the chances of going into a death trade is almost 0. If you reenter every 10 points you can probably surmise of very little chance of facing the death trade over the year, and that is assuming it did 150 points without retracing, a very rare event in currency trading. So here is the info and use it to develop your strategy. I think you will be surprised that some currencies do fit together and could use the same formula successfully. I color coded some of the groupings for an easier read. Happy trading.

Bob
thanks a lot for 7 year currency facts.xls
  • #256
  • Feb 29, 2008 8:44am Feb 29, 2008 8:44am
Quoting effilang
http://www.forexfactory.com/attachme...1&d=1197313611



looooooool - best picture EVER

If he would have sold the house then his account would have been big enough to take the drawdown.
  • #257
  • Feb 29, 2008 10:54am Feb 29, 2008 10:54am
Sorry for the belated response. Just catching up with this thread again now.
Quoting nanningbob
I said I would give some pointers about avoiding the death trade.......

Bob, many thanks for this. Will have a think about it.
_________________________________________________
Quoting jnevins
......In a heads/tails system, there is much confusion regarding future expectancy. It seems to be a gamblers expectation that with each heads or tails comes a greater likelihood that the opposite should come up. In other words, after hitting 6 heads in a row, there is increasing expectancy that the next throw will come up tails.... The question I am asking.... is that true? Is there any math that would back up this assertion or prove the opposite... .......

Jerry, there is always a 50/50 chance of heads or tails, at any given point. To think otherwise is known as "gambler’s fallacy".

Explained further here: http://www.forexfactory.com/showthread.php?p=1786938
See also: http://en.wikipedia.org/wiki/Gamblers_fallacy
More advanced material here: http://mathworld.wolfram.com/CoinTossing.html

David
  • #258
  • Feb 29, 2008 11:55am Feb 29, 2008 11:55am
Quoting Peanutdude
If he would have sold the house then his account would have been big enough to take the drawdown.
Better yet would have been to take your profit out every week, then you would always be playing with free money. There is always risk involved in every trading system, its how you manage it. To some, martingale will always look fool hardy, but to those who manage and understand it, it can be a gold mine. Anyways I am enjoying 6 weeks of profits and have had only 4 losing days. One which was for -.22 cents. I am now playing with free money meaning if I wipe out both accounts I am still in profit because I take my profits out every week. Happy trading
  • #259
  • Mar 1, 2008 9:23am Mar 1, 2008 9:23am
Quoting hanover
Sorry for the belated response. Just catching up with this thread again now.

Bob, many thanks for this. Will have a think about it.
_________________________________________________

Jerry, there is always a 50/50 chance of heads or tails, at any given point. To think otherwise is known as "gambler’s fallacy".

Explained further here: http://www.forexfactory.com/showthread.php?p=1786938
See also: http://en.wikipedia.org/wiki/Gamblers_fallacy
More advanced material here: http://mathworld.wolfram.com/CoinTossing.html

David
Thanks David! That is just the information I was looking for. I was researching Martingale theory and its history back to Provence in the 18th century. The entire idea is predicated on what probability is there that you could go x levels before a 50/50 event MUST go in your favor. The Fallacy is that after 4 tosses of a heads, there is now a 1 in 32 chance that the 5th toss will be a heads... That is wrong there is still a 50/50 chance with a fair coin. There is a psychological explanation in the Wiki article...
"Some claim that the gambler's fallacy is a cognitive bias produced by a psychological heuristic called the representativeness heuristic, and a related phenomenon called pareidolia[attribution needed]. There is an argument that we are programmed to look for patterns in chaos ("Is that a tiger half-hidden in the trees?" "Is that a bunch of ripe fruit half-hidden in the leaves?") and are actually biased towards spotting patterns when none exist. An animal that is prone to over-imagining patterns (e.g., never misses real tigers, but sometimes sees imaginary ones) is far more likely to pass on its genes than a cousin which ignores just one real tiger."

This explains much in the Forex world as well! Thanks Bob and all contributors to this thread.
  • #260
  • Mar 2, 2008 1:27am Mar 2, 2008 1:27am
Quoting jnevins
Thanks David! That is just the information I was looking for. I was researching Martingale theory and its history back to Provence in the 18th century. The entire idea is predicated on what probability is there that you could go x levels before a 50/50 event MUST go in your favor. The Fallacy is that after 4 tosses of a heads, there is now a 1 in 32 chance that the 5th toss will be a heads... That is wrong there is still a 50/50 chance with a fair coin. There is a psychological explanation in the Wiki article...
"Some claim that the gambler's fallacy is a cognitive bias produced by a psychological heuristic called the representativeness heuristic, and a related phenomenon called pareidolia[attribution needed]. There is an argument that we are programmed to look for patterns in chaos ("Is that a tiger half-hidden in the trees?" "Is that a bunch of ripe fruit half-hidden in the leaves?") and are actually biased towards spotting patterns when none exist. An animal that is prone to over-imagining patterns (e.g., never misses real tigers, but sometimes sees imaginary ones) is far more likely to pass on its genes than a cousin which ignores just one real tiger."

This explains much in the Forex world as well! Thanks Bob and all contributors to this thread.
50-50 argument with coins is valid because the choice is limited to only two possibilities. No matter how many times you flip there is always a 50-50 chance of getting heads or tails. However, that is not the same in other areas. For instance you have a deck of cards which is 52 possible combinations multiplied. If you are playing 21 and the first two cards are not a face card or an ace you have increased the odds of getting it the next time, even though you have only increased the odds a small increment. Lets say after 26 cards you still have not shown any face cards,10 or ace; 20 of the last 26 cards are now either an ace, face card or a 10, now you have an excellent chance of getting 21 in the last 13 hands even though it is possible of getting many 20s or two aces together in one hand. The same thing is true in Forex. I have never seen or heard of a trading day where price never retraced somewhere during the day. You can correct me if I am wrong, and even if it did happen one time, I can safely and accurately predict 100% of the time that during a trading day price will retrace. With this fact in mind, it has nothing to do with coin flips because we are not dealing with a 50/50 possibility but a 100% chance probability. The only decision left is what parameters can I put into my EA that will keep me in this 100% probability range, meaning at one point do I think price will not keep going on in one direction forever. This more accurately portrays what we are trying to do, forex is far more complicated than a coin flip and is much more reliable in retracing than heads coming up 50 times in a row. I believe with 100% certainty that price will retrace on any given day, the question is how far will the GBP/USD go before that is possible? Can I play that certain possibility within a given range? And what are my chances price would ever go beyond that range and not retrace? The greatest move for GBP/USD on one day (high/low) was 441 pips in the last 7 years. That day was Feb. 20, 2004 the question is during the biggest day, was there any retracements of over 50 points during that run so I could profit if I was entered against the trend? The answer is yes there were several. If I was using sto as my indicator to enter trades that day would I have been in trouble? The answer is a big yes. I would have sat at the over sold area all day with it entering buys with every retracement, when you really wanted a sell. If I use a MA cross would I have been in trouble? NO, because once I collected my first profit I would have been with the trend the rest of the day. These are just my thoughts.

Bob
2 traders viewing now