Is This My Holy Grail EA :D | Page 12 | 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

  • #221
  • Edited 9:01am Feb 18, 2008 8:49am | Edited 9:01am
Quoting hanover
Bob, if the top traders say that, at the time of entry, they can't be 100% sure which trades will be winners, what makes you so confident? You may be right, of course, I'm just curious to know how you can be certain.

If you're 100% certain upon entering a trade that the trade will be a winner, then you should be sizing your position to risk as much as possible of your entire account on that trade. If you're not willing to do so, then I'd suggest that you're somewhat less than 100% certain.

Anwyay, if you have a positive expectancy system, why risk the death trade at all? It can be shown mathematically that Martingale (or anti-Martingale, or in fact any system that varies bet size) fares no better than flat betting, in any activity that involves uncertainty (see my three consecutive posts in the thread, starting here: http://www.forexfactory.com/showthread.php?p=198995). Increase in bet size is justified only when the probability of success can be shown statistically to be higher than "normal". If the Martingale MM is - or at least appears to be - the indispensable component that is making your system profitable, then I'm afraid the method behind the entries and exits is flawed.

Anyway, I sincerely wish you all the best. I'm not trying to be a naysayer, just trying to save your bankroll (and possibly others' bankrolls also).

David

[Edit:] more info on MM (among other things) in my post here: http://www.forexfactory.com/showthread.php?p=1848426
MM in itself is not the answer.

You may also find the table at the end of the following post useful, for determining the probability of losing trade(s): http://www.forexfactory.com/showthread.php?p=188593
Thank you for your post and your well thought out comments. I do not look at any trade being 100%, I don't think there is such a thing. The best you can do is stack the odds in your favor and pull the trigger, have some way to protect your equity and go on. If I am going to give you a close to 100% system I would say this. After trading 1 year, I decided that I would try to find what was 100% true as possible. In other words what is always true when it comes to forex trading. These were my conclusions. Price is always moving up and down it is never static. I threw out the idea that price is in consolidation mode or trending mode, instead I viewed it as price moves up in down in greater or lesser strength. It is either going up or down, it never stays put. The second conclusion was there is always a high and a low. Every 1m 5m 15m 1h etc. has a high and low. The third conclusion was price always changes direction in every time period. (maybe in a 1M chart it is 1 pip) These three things are always 100% true. So the Martingale always takes advantage of the fact that price will always retrace. What goes up will come down, that is 100% true in Forex trading. Thats when you can make a profit, or if you are trading with the trend it always makes money. I once did a test with a Azam's Martingale. I had it on always buy. No matter which direction price went it would always win with one exception. The killer to a martingale system is a trend that doesnt retrace and that usually happens at big news stories. Stay away from them and I can trade with certainty. When I turn on the EA I walk away and come back several hours later and see how much money it has made me. So I do trade with certainty. The only thing I have to fear is the big news story. I turn off my EA during those times. I hope this is clear and understandable.
  • #222
  • Feb 18, 2008 12:19pm Feb 18, 2008 12:19pm
Quoting nanningbob
Thank you for your post and your well thought out comments. I do not look at any trade being 100%, I don't think there is such a thing. The best you can do is stack the odds in your favor and pull the trigger, have some way to protect your equity and go on.......<excellent point, I agree 100%>
Bob, many thanks for sharing your ideas. Let's make sure that I understand correctly. You're saying that:
(1) apart from the situation where there's a big news story, it's inevitable that price will eventually retrace to a mean; and
(2) your testing has shown that the Martingale will always deliver the one precious win that returns your account to its previous high, before the death trade has the opportunity to occur.
I agree that, IF you can say this with 100% certainty, then you have an infallible system.

I think it's dangerous to assume that back-testing can promise 100% certainty, since the number of situations tested can never be infinite. With only a finite number of events, we can perhaps talk about levels of statistical confidence, but never complete certainty. Moreover, price movement is driven by emotions like fear and greed, so unlike dice or card games, we can never calculate probabilities with perfect accuracy.

If you're timing your entries and exits in such a way to exploit point #1, then you could well have a positive expectancy method. (In which case you're probably ahead of where I'm currently at, in my own journey). But even with a positive expectancy method, your long term profit expectancy is exactly the same whether you use Martingale or flat bet size (see the first link in my previous post for the mathematical proof). But, with anything less than an infallible system, the Martingale introduces the possibility of the death trade. So the Martingale increases risk, for no additional benefit. This is true regardless of how the entries and exits operate.

I think people like the idea of the Martingale because losses are hard to stomach, and the Martingale promises a total recovery with just one win...... but with the proviso that one has infinitely deep pockets (given enough time, even Bill Gates would go broke). Flat betting will also recover our account balance, it just takes a lot longer.

David
  • #223
  • Feb 18, 2008 11:54pm Feb 18, 2008 11:54pm
Quoting hanover
Bob, many thanks for sharing your ideas. Let's make sure that I understand correctly. You're saying that:
(1) apart from the situation where there's a big news story, it's inevitable that price will eventually retrace to a mean; and
(2) your testing has shown that the Martingale will always deliver the one precious win that returns your account to its previous high, before the death trade has the opportunity to occur.
I agree that, IF you can say this with 100% certainty, then you have an infallible system.

I think it's dangerous to assume that back-testing can promise 100% certainty, since the number of situations tested can never be infinite. With only a finite number of events, we can perhaps talk about levels of statistical confidence, but never complete certainty. Moreover, price movement is driven by emotions like fear and greed, so unlike dice or card games, we can never calculate probabilities with perfect accuracy.

If you're timing your entries and exits in such a way to exploit point #1, then you could well have a positive expectancy method. (In which case you're probably ahead of where I'm currently at, in my own journey). But even with a positive expectancy method, your long term profit expectancy is exactly the same whether you use Martingale or flat bet size (see the first link in my previous post for the mathematical proof). But, with anything less than an infallible system, the Martingale introduces the possibility of the death trade. So the Martingale increases risk, for no additional benefit. This is true regardless of how the entries and exits operate.

I think people like the idea of the Martingale because losses are hard to stomach, and the Martingale promises a total recovery with just one win...... but with the proviso that one has infinitely deep pockets (given enough time, even Bill Gates would go broke). Flat betting will also recover our account balance, it just takes a lot longer.

David
Well, there is no 100% certainty. We both know there is no such thing. I have been studying the Martingale system on and off for a year. If you read my posts you know I clearly recognize the death trade. Its not avoiding it, it is managing it. Managing the Martingale runs against 90% of the prevailing opinions of the day. You dont use stop loss, you dont use fundamentals, techinical analysis is of little use even though we have to use an indicator to start the trade. It is not based on anything the ebooks or courses taught in the world, it is based on one simple fact, the market moves up and down and you can stack the odds in your favor using this action. That is what all currency trading is, using some method that stacks the odds in your favor and you pull the trigger. The martingale uses normal market movement and puts it in your favor. The whole Fibonocci method of trading is based on retracements. If the market moves X number of pips it will retrace a certain percentage back. Pivot point method is: the market moves to a certain resistance or support point and bounces back. Stochastics method when price goes to oversold or overbought look for the retracement or change of direction. News trading, if the news is above predictions look for this many buy pips, if it is below expectations look for sell pips. Every method uses this to some form. Here are the odds, practically every single day the Eur/USD moves 75-100 points unless there is a big news story or govt. action. I can play for 15 point profit and go 12 levels deep or 180 points without worrying about the death trade. The price can push close to 200+ points before I have to worry about wiping out my account. How many days last year did the Euro move over 200 points in one day?? Well there were 8 days and that is assuming it never retraced 15 pips during that run. So if I set up my numbers to accomodate up to 200 pips in a martingale trade I am looking at only 8 days out of the year of wiping out my account in the death trade. I like those odds much better than any other method. The difference is Azam and company have coded an EA that makes the Martingale system work beautifully.

The martingale method flies in the face of prevailing methods and currency trading philosophy. The two problems I face now is long term live testing and what figures work best with what currencies. I will admit I do not have all the answers yet, but this is closer than any other system I have tried. My advice is this, Demo live until you understand how it works. Live trade with pennies to get the feel of trading with real money. (that is where I am now, I will commit 2-400 dollars) Taking my kids to a Magic games costs that much. Once I fully understand this and am comfortable with it I will then try nickels or multiple accounts. So far I am ahead of the game and hope to continue to do so because I pull the money out every week. Its like getting a paycheck.
  • #224
  • Feb 19, 2008 5:55am Feb 19, 2008 5:55am
Bob,
Excelent explanation about the system. I won´t say better.
  • #225
  • Edited 6:21pm Feb 19, 2008 5:47pm | Edited 6:21pm
Quoting nanningbob
Well, there is no 100% certainty. We both know there is no such thing.....
Bob, many thanks for your very detailed explanation.

If I understand correctly, you’ve allowed enough leeway in your account to allow losses (and hence doubling up) to level 12, which can only happen whenever the Euro moves more than 180 pips in a day, without an intervening 15 pip retracement. Statistically, the probability of this occurring is very low (significantly less than 8 days in each year). So you are guaranteed a very small profit (percentage-wise) every day…… provided that a loss at level 12 never occurs.

I’m in no way being critical of this approach, but personally I don’t have the stomach to trade using a Martingale-based system.

I stand by what I said in my earlier posts. Mathematically your expectancy is the same, whether you double the position size at each level, or not. Any "edge" lies in the extent to which the Euro is more likely to revert to a mean, during the day, the further that it moves away from one.

I wish you all the best with your long-term live testing.

David
  • #226
  • Feb 20, 2008 5:56am Feb 20, 2008 5:56am
Quoting hanover
Bob, many thanks for your very detailed explanation.

If I understand correctly, you’ve allowed enough leeway in your account to allow losses (and hence doubling up) to level 12, which can only happen whenever the Euro moves more than 180 pips in a day, without an intervening 15 pip retracement. Statistically, the probability of this occurring is very low (significantly less than 8 days in each year). So you are guaranteed a very small profit (percentage-wise) every day…… provided that a loss at level 12 never occurs.

I’m in no way being critical of this approach, but personally I don’t have the stomach to trade using a Martingale-based system.

I stand by what I said in my earlier posts. Mathematically your expectancy is the same, whether you double the position size at each level, or not. Any "edge" lies in the extent to which the Euro is more likely to revert to a mean, during the day, the further that it moves away from one.

I wish you all the best with your long-term live testing.

David
Dave I would like to invite you to keep posting and making comments. I learn more from people who disagree than those who agree sometimes so keep firing away. I would also like to note that some of us have learned a secret to the Martingale and that is to not double up but to use different figures and reentry points. Even though we have not made it a complete science yet, it greatly changes the odds of the martingale system. It is similiar but it is not he same. When I say the most important thing you can do is manage the death trade that is one of the key components. There are 3 or 4 that I have discovered and that helps me to prevent the death trade. When I have a little more time I would like to write about them in the near future. Basically, if you eliminate the death trade your success rate goes way up and it isnt as difficult as it may sound. But it does take some practice and know how. Good luck.

Bob
  • #227
  • Feb 20, 2008 5:39pm Feb 20, 2008 5:39pm
Quoting nanningbob
Dave I would like to invite you to keep posting and making comments.....
Hi Bob,

As you point out, the key is to get as close as possible to eliminating the death trade. If it were (hypothetically) possible to eliminate its possibility altogether, then you’d have the perfect forex system. Anyway, if you're willing to share these ideas, I’d be very interested.

I remember reading in a thread, a few months ago, how somebody used a "Fibonacci martingale". Instead of doubling his bet size (1,2,4,8,16,32,64,128,256,512,1024,2048, etc), he used Fibo numbers (1,2,3,5,8,13,21,34,55,89,144,233, etc). As we can see, after 12 levels, the stake is only 233 x the original bet, instead of 2048 x, i.e. significantly less risk. The compromise that I see is this. With the original martingale, it only takes one win to recover all previous losses; with the Fibo version it requires multiple wins. In other words, by reducing the size of the death trade, we are making recovery more difficult. Again, it illustrates the delicate compromise that exists between return and risk.

Earlier in the thread somebody discussed periodically taking profits from the account. Again, I think there’s a compromise involved. Removing funds reduces the number of levels one can safely play, thereby increasing the probability of the death trade.

Just to confirm what I mean by expectancy. If we double up, then after 12 failures our loss is 1+2+4+8+16+32+64+128+256+512+1024+2048 = 4095. On a winning day, we are going to win +1, provided we managed at least one win at any level. On a losing day (the death trade) our loss is –4095. So provided a loss occurs, on average, no more often than every 4096 days, we have a positive expectancy system. But if a loss occurs, on average, every 4095 days or less, then we will end up losing money in the long term.

So if, for example, we’re doubling up to a maximum of 12 levels each day, and there are 250 trading days in a year, it would only take one death trade in every 4095 / 250 = 16.4 years to wipe us out. That’s one major reason why I said I wouldn’t have the stomach to trade in this way. It also means that any forward testing would ideally need to take a very long time.

You’ve highlighted the two parameters you have to work with: (1) bet size, and (2) timing of entry. As I’ve already said, I don’t believe that varying bet size can improve expectancy**. But if we could somehow time entries around higher probability reversal areas (support/resistance, perhaps?), then I believe expectancy can be improved. (But I expect it would be difficult to code ideas like S/R into an EA). Anyway, from what you’ve said, I gather you already have ideas on this.

Let’s try to use a Roulette analogy, where the 12 levels are represented by 12 consecutive spins of the wheel. A black number is a winning trade, a red (or a zero, simulating the spread cost) is a loss. In a ‘normal’ game of Roulette, we can show that it irrespective of whatever betting system we use, nobody can ever beat the casino in the long term. But, if we can show that the more the Euro moves away from a mean price during the day, the greater the likelihood of a reversal, then we would need to adjust our analogy accordingly. After each red is spun, we progressively remove red numbers from the wheel, making the probability of spinning black increasingly higher. Now we are starting to improve expectancy. That, I believe, is a real "edge".

If (and I stress IF) the more the Euro moves away from a mean price during the day, the greater the likelihood of a reversal, then the further the move, the more expectancy increases. So why not wait until a move of (say) 100 pips, before you decide to open your level 1 trade? Since moves of >180 pips are extremely rare, you’ve both (1) increased the probability of a successful trade, and (2) significantly reduced the effect of the death trade (since you won’t need so many levels). If the Euro doesn’t move 100 pips, then you simply don’t trade on that day.

Another possibility - that might improve expectancy - would be to only take trades in the direction of the long term trend.

Hopefully I’ve given you some more ideas to work with.

David

** PS – Please check out the attached XLS. (It’s an enhancement of one that was included in one of my previous links). It allows you to try out any betting system, and playing strategy, over 1–6 trials (trades). Convince yourself that no matter how many trials, no matter how/when you decide to quit, and no matter how you vary your bet sizes, it makes absolutely no difference to the overall result. Then you need to decide how your trading approach differs from this, i.e. precisely where it provides an edge that goes beyond the mathematical theory.
Attached File(s)
File Type: xls Martingale comparison testing.xls   157 KB | 724 downloads
  • #228
  • Feb 21, 2008 12:19pm Feb 21, 2008 12:19pm
Change of thought. Redo the Martingale and modify it at 1.5x instead of 2.0x now your levels are 1,2,3,5,8,12,18,27,41,62,93,1.40 (these numbers come from actual live demo trades on InterbankFX using .01 trades) you are now 12 levels deep. The increase in funds in your account aren't the usual ridiculous numbers of most martingale backtests but are steady and regular. If your reentry is 25 points (4 levels per 100 pips) that puts you in play for 300 points everyday without the death trade and still you have room to maneuver. For instance in the last 7 years the following currencies biggest one day move is: EUR/GBP -193 EUR/USD -319 EUR/CHF 334 CHF/JPY 335 NZD/USD - 345 CAD/USD - 379 AUD/USD - 382 GBP/USD - 441 AND USD/JPY - 474. Using 1.5 in different forms allows an increase in your account but is very close to eliminating the death trade in 9 currencies. Even if the second day of trading continues in the same direction without a retrace, you have plenty of time to exit a bad trade. Time and price action are now on your side as far as statistics go. Except for the GBP/USD and the USD/JPY, 7 of these currencies only went above 300 pips in one day a total of 15 days in 7 years. To be honest with you I dont know any system that is this close to being worry free. Now these figures I have given you I figured out yesterday so I am now writing some code that uses these numbers and testing them out. Imagine in over 2000 trading days I have only 15 days that had the potential of damaging my account in 7 currency groups. I think this is exciting but so far I want to say "unproven in live trading." I think this has significant possibilities. I will keep you informed. Bob
  • #229
  • Feb 21, 2008 4:06pm Feb 21, 2008 4:06pm
Quoting nanningbob
Change of thought. Redo the Martingale and modify it at 1.5x instead of 2.0x now your levels are 1,2,3,5,8,12,18,27,41,62,93,1.40 (these numbers come from actual live demo trades on InterbankFX using .01 trades) you are now 12 levels deep. The increase in funds in your account aren't the usual ridiculous numbers of most martingale backtests but are steady and regular. If your reentry is 25 points (4 levels per 100 pips) that puts you in play for 300 points everyday without the death trade and still you have room to maneuver. For instance in the last 7 years the following currencies biggest one day move is: EUR/GBP -193 EUR/USD -319 EUR/CHF 334 CHF/JPY 335 NZD/USD - 345 CAD/USD - 379 AUD/USD - 382 GBP/USD - 441 AND USD/JPY - 474. Using 1.5 in different forms allows an increase in your account but is very close to eliminating the death trade in 9 currencies. Even if the second day of trading continues in the same direction without a retrace, you have plenty of time to exit a bad trade. Time and price action are now on your side as far as statistics go. Except for the GBP/USD and the USD/JPY, 7 of these currencies only went above 300 pips in one day a total of 15 days in 7 years. To be honest with you I dont know any system that is this close to being worry free. Now these figures I have given you I figured out yesterday so I am now writing some code that uses these numbers and testing them out. Imagine in over 2000 trading days I have only 15 days that had the potential of damaging my account in 7 currency groups. I think this is exciting but so far I want to say "unproven in live trading." I think this has significant possibilities. I will keep you informed. Bob
Hi my friend Bob.

So, if you use 1.5x factor, you will need more than 1 level retrace correct... maybe 2 ?

1.5x, 7 leves:
0.10, 0.15, 0.30, 0.45, 0.67, 1,05, 1,57

20 pips tp, and 10 pipstep level.

start at 1.2000 at maximum it goes to 1.2060 (60 pips).. and start retrace and Close at 1.2040 (40 pips) distance.
1.2000 :0.10 = - 40 pips = - $4.00
1.2010 :0.15 = - 30 pips = - $4,50
1.2020 :0.30 = - 20 pips = - $6,00
1.2030 :0.45 = - 10 pips = - $4,50
1.2040 :0.67 = 0
1.2050 :1,05 = + 10 pips = + $10,50
1.2060 :1,57 = + 20 pips = + $31,40

Result: $ 22,90

If u you use a good ratio, from TP and Pipstep, it´s really a good choice.
  • #230
  • Feb 21, 2008 8:14pm Feb 21, 2008 8:14pm
Quoting Numeral
Hi my friend Bob.

So, if you use 1.5x factor, you will need more than 1 level retrace correct... maybe 2 ?

1.5x, 7 leves:
0.10, 0.15, 0.30, 0.45, 0.67, 1,05, 1,57

20 pips tp, and 10 pipstep level.

start at 1.2000 at maximum it goes to 1.2060 (60 pips).. and start retrace and Close at 1.2040 (40 pips) distance.
1.2000 :0.10 = - 40 pips = - $4.00
1.2010 :0.15 = - 30 pips = - $4,50
1.2020 :0.30 = - 20 pips = - $6,00
1.2030 :0.45 = - 10 pips = - $4,50
1.2040 :0.67 = 0
1.2050 :1,05 = + 10 pips = + $10,50
1.2060 :1,57 = + 20 pips = + $31,40

Result: $ 22,90

If u you use a good ratio, from TP and Pipstep, it´s really a good choice.
The figures you use would work well on the Eur/gbp I think. Other currencies you would have to worry about news stories. If you use your formula during non news stories I think you would be in good shape. The only thing left to calculate is the amount of funds you need so you dont crash out. My next post which I will write later will be about something written in Azam's code that will stop the death trade. Catch you later.
  • #231
  • Feb 21, 2008 8:16pm Feb 21, 2008 8:16pm
yes 2 levels works but 3 taking profit is the way to go. That thinking sends you on the right track.
  • #232
  • Feb 22, 2008 11:40am Feb 22, 2008 11:40am
I said I would give some pointers about avoiding the death trade:
1. stay away from news stories especially the big ones unless you have developed a system that will handle 100-200 pip moves that can happen at news time. I am working on one for the gbp/jpy that is a 100 take profit and can go 5-600 pips against you without wiping out your account. Wish me luck on that one.
2. trade the currencies that dont move as much, any USD except the GBP/USD, any of the euro currency pairs but be wary of the Eur/jpy
3. In the code of Azam's EA there are two account management choices. One is a percentage of the account. If it goes more than, lets say 20% it closes all trades, HOWEVER, it will reopen new positions and you have less equity.
4. I like this one the best. It allows you to set a number for to close all trades. For example if you have 500 in your account and you dont want to lose more than 150 you set it at 350. Then you dont have to worry about it starting new trades. You can come back later and set up new trade parameters. This is good for those that want to have larger sums and use the Martingale system. You could use 2000 dollars and set it at 1800 and limit your losses to 200 dollars max.
5. Make sure you have enough dollars to cover 12-15 levels deep on a trade. DO the MATH. It is frustrating to see a trade go back in your favor but you missed it because you did not have enough funds to carry the trade.
6. You lose money when you get greedy. I gave my friend one of the EAs I had consistantly made money on. I told him to demo for a couple of weeks to get used to it and then use .01 trades till he really understood the EA. Well, he didnt listen. He put 2500 dollars in an account, put it on .05 cents and went right into the Euro and US opens disregarding my warnings about news stories. He lost 1700 dollars in 9 hours. If he had done penny trades he would have made 60 bucks that day but he didnt listen or follow the rules. 1-2% earnings a day is a very good investment, dont forget that.
7. When you get past your last level and it starts going more against you, you are heading for the death trade. Find a good exit and get out and take the loss. You can get your money back in 2-3 days if you exit but you will never get it back if you let the death trade go all the way to its end.
8. Dont ever hold a trade over the weekend because price can gap against you by 100 pips or more. Long term trading this is not as important but with a Martingale system it could be suicide.

I hope this helps.

REMEMBER, You can successfully manage the death trade but it takes practice and experience. You cant do a Martingale system successfully without putting the training time in.
  • #233
  • Feb 22, 2008 11:46am Feb 22, 2008 11:46am
I forgot one more thing. you can limit your losses and the death trade by controling the number of levels you use. I use 9-12 levels depending on the amount of money I have in my account. The less money the fewer levels I use. I make sure the multiplying effect doesnt get so big that it would clean out my account. DO THE MATH
  • #234
  • Feb 23, 2008 11:52am Feb 23, 2008 11:52am
Bob,

Thanks so much for such a great series off posts. I think one should design the EA on a stochastic system or one that is totally random. 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... that the Heads/Tails throw is truly random and that the likely hood of going 11 levels or even 16 levels is a known quantity. How many levels does one have to go to approach infallibility... say 1 in 10,000 or 1 in 100,000 chances that you will need to go X levels and have a meltdown?

-Jerry
  • #235
  • Feb 24, 2008 1:32am Feb 24, 2008 1:32am
Quoting jnevins
Bob,

Thanks so much for such a great series off posts. I think one should design the EA on a stochastic system or one that is totally random. 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... that the Heads/Tails throw is truly random and that the likely hood of going 11 levels or even 16 levels is a known quantity. How many levels does one have to go to approach infallibility... say 1 in 10,000 or 1 in 100,000 chances that you will need to go X levels and have a meltdown?

-Jerry
I do not know the odds for coin flips but here are the odds for certain currencies over the last 7 years.
If your reentry is 25 points (4 levels per 100 pips) that puts you in play for 300 points everyday without the death trade and still you have room to maneuver. For instance in the last 7 years the following currencies biggest one day move is: EUR/GBP -193 EUR/USD -319 EUR/CHF 334 CHF/JPY 335 NZD/USD - 345 CAD/USD - 379 AUD/USD - 382 GBP/USD - 441 AND USD/JPY - 474. Using 1.5 in different forms allows an increase in your account but is very close to eliminating the death trade in 9 currencies.

Times going over 300 pips in one day in the last 7 years: eur/gbp 0 times, eur/usd 2 times, eur/chf 1 times, chf/jpy 2 times, nzd/usd 2 times, cad/usd 3 times, aud/usd 3 times, aud/nzd 2 times, gbp/usd 27 times, usd/jpy 10 times, usd/chf 44 times, eur/jpy 42 times, nzd/jpy 22 times, aud/jpy 25 times, gbp/jpy 210 times, gbp/chf 137 times. This is over a possible 2151 trading days. The couple hours trading on Sunday night is included as a full trading day so that is 6 possible trading days a week. You can do the math on each pair to get your percentages. I have attached my research as word document. Also, when you calculate how much funds you need the eur/gbp is almost 2 x the funds needed per us dollar compared to the eur/usd and many of the Japanese funds are less per dollar.
Attached File(s)
File Type: doc Daily spreads highs 7 year period.doc   24 KB | 633 downloads
  • #236
  • Feb 24, 2008 1:46am Feb 24, 2008 1:46am
The percentages would be for 1 trading day. This does not preclude the possibility of a currency going two or three days in a row without the retrace happening. I consider that scenario highly unlikely but possible. I would assume you would cut your losses and drop out rather than letting the loss run another day. Lets say your 12th level entered at 2.0000 dollars and your take profit would be 1.9975. It never kicks in the day of your trade. If the second day price kept rising and never went back to 1.9975 you could wipe out your funds even though price may only move 100 points that one day. As a general rule with Martingale system you dont carry your trades that are many levels deep into a new day. You should cut your losses at some point. I have made myself a rule using pivot points. If I go behind my last level I look at my next pivot level (resistance or support level) and if it breaks through that level I am out. I had one losing trade that bounced off the next level (resistance level) came back down and I cut my loss in half rather than taking it immediately. Knowing some technical analysis or fundamental analysis is helpful in making the decision on taking a loss. Another thing I have done once is I saw a take profit that went 5 points past a support level and I changed it to 3 points above that level. I have only done it once mainly because it was a 10 level entry and I didnt want to just miss the take profit because of a bounce off a support line.
  • #237
  • Feb 24, 2008 1:52am Feb 24, 2008 1:52am
I am also comparing on live demo trades comparing using a sto system versus a moving average system. The weaknesses of both are the following. When the sto goes way overbought or way over sold, you are entering trades in the opposite direction but price can still keep going. The sto is a bad indicator when price is running on a currency. The MA weakness is whenever there is a price turnaround it makes an entry at the peak and then price goes the opposite direction. So that is my observations using those two indicators. I am beginning to prefer the MA because when price turns around there usually is a battle between the bears and bulls and you get your TP within the range you have set up. I use 4 hours charts to predict my trend using the MA indicator.
  • #238
  • Feb 24, 2008 2:03am Feb 24, 2008 2:03am
To figure odds per trading day I will take the most volatile currency.
gbp/jpy went over 210 times above 300 points in one day. This is over a possible 2151 trading days so your odds are 10% or 1 day in 10 you could face the death trade. Staying away from big news days or only trading in off hours increases the odds in your favor with this pair and you may be able to safely trade it. This is an example of how you would want to trade this pair.
  • #239
  • Feb 24, 2008 6:22pm Feb 24, 2008 6:22pm
Quoting nanningbob
To figure odds per trading day I will take the most volatile currency.
gbp/jpy went over 210 times above 300 points in one day. This is over a possible 2151 trading days so your odds are 10% or 1 day in 10 you could face the death trade. Staying away from big news days or only trading in off hours increases the odds in your favor with this pair and you may be able to safely trade it. This is an example of how you would want to trade this pair.
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?
  • #240
  • Feb 25, 2008 8:00am Feb 25, 2008 8:00am
Quoting nanningbob
The percentages would be for 1 trading day. This does not preclude the possibility of a currency going two or three days in a row without the retrace happening. I consider that scenario highly unlikely but possible. I would assume you would cut your losses and drop out rather than letting the loss run another day. Lets say your 12th level entered at 2.0000 dollars and your take profit would be 1.9975. It never kicks in the day of your trade. If the second day price kept rising and never went back to 1.9975 you could wipe out your funds even though price may only move 100 points that one day. As a general rule with Martingale system you dont carry your trades that are many levels deep into a new day. You should cut your losses at some point. I have made myself a rule using pivot points. If I go behind my last level I look at my next pivot level (resistance or support level) and if it breaks through that level I am out. I had one losing trade that bounced off the next level (resistance level) came back down and I cut my loss in half rather than taking it immediately. Knowing some technical analysis or fundamental analysis is helpful in making the decision on taking a loss. Another thing I have done once is I saw a take profit that went 5 points past a support level and I changed it to 3 points above that level. I have only done it once mainly because it was a 10 level entry and I didnt want to just miss the take profit because of a bounce off a support line.
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 ?
0 traders viewing now