This so called "self-destruction" of trading systems... | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Top 100 best Trading Systems: post good Trading Systems here 62 replies

Broker Challenge "NDD/STP" vs. True "ECN" and "Mkt Mkr" broker talk 46 replies

Dealing with "Invalid Price", "Requote", "Server Busy" in MT4 3 replies

Are the best systems also the systems that fail most spectacularly? 55 replies

Systems, Systems and Systems... 1 reply

This so called "self-destruction" of trading systems...

  • #41
  • Mar 16, 2007 9:07am Mar 16, 2007 9:07am
Quoting ilanr
For that matter, you surely had read Edwards & Magee before you formulated your method. ]
Have to admit that I got this book first time in my hands only after the method had been fully developed. Before that I have read only Murphy's book, what then allowed me to test and add some technical templates to the bunch. But the initial idea about using ATR came to me just from own practical observations.
  • #42
  • Mar 16, 2007 9:26am Mar 16, 2007 9:26am
Quoting ilanr
I have no idea, what would happen if several big players would all trade the same system all of the time. Maybe, indeed, the system will stop working. However, if the microcosmos of FF and other fx forums reflects the psychology of traders in general, there is no slightest chance that any significant group of traders will follow exactly the same system.

I'm fascinated with this psychology (I'm part of it myself, of course). Dozens of profitable systems were published at FF. Even the most severely sub-optimized systems, like the Pouria method that has attracted lot of attention recently. Where are all the Pouria traders? They moved on to some new fades. If they would have taken some time to optimize Pouria, they could make some nice profits. I optimized it, I saw it could work. Why, right after that, did I start working on another system? Well, trading doesn't leave me much time to go to my shrink. If I'd see him, I'd certainly like to discuss this question with him.

Publish any system you want. Several thousand people will read your post. Several hundred will try it. Several dozen will suggest adding to it any possible indicator. Zero will stick to this system after some period of time. This probably happens with 95% of published systems. The remaining 5% remain alive mostly because of leadership skills of their authors, not because they are any better or worse than other systems.

This behavior reminds me of restaurant-going behavior in my hometown of Tel-Aviv. A new restaurant/bar/cafe is opened. Everyone is talking about it. Queues at the entrance. Newspaper critics hail it. Half a year passes. Nobody remembers the place. Everyone is in the queue to the next culinary Holy Grail. This is bad for restaurant keepers. This is great for food-lovers: you can always find a great place to eat without ordering a table a week ahead.
This was a good post. It is interesting how people operate isnt it? I did it myself for about 4-5 months. I would get a system and use it for a week or two.. then try another. Though, I think this is part of the learning phase ... not a problem. If a trendline system is the first thing a person sees and they are going through growing pains with it and then they see an MA system or BB system... they immediately assume that is what they are missing. By going through all these systems they will become used to all those indicators. I do think after a while it is important for a person to basically write up a system themselves. I personally have been more profitable in the last 2 1/2 months since I sat down, put indicators on a chart and told myself I would not stray from them. I will give it until the end of this year to decide.

Another point about systems. Jimmy Young has done some free presentations where he presented using a 60SMA on a chart. Well, I know a trader that uses that along with basic trendlines. The thing is... he has presented using a 60SMA to thousands of people. How many you suppose use it? I know I dont... why not? This person I know is profitable with it? Maybe I need to see that shrink.
The market is my nation. Traders, my family. Hello, brothers and sisters!
  • #43
  • Mar 16, 2007 9:28am Mar 16, 2007 9:28am
Quoting Igrok
Have to admit that I got this book first time in my hands only after the method had been fully developed. Before that I have read only Murphy's book, what then allowed me to test and add some technical templates to the bunch. But the initial idea about using ATR came to me just from own practical observations.
So, you probably enjoyed developing your method even more than I thought you did!
sans peur et sans reproche
  • #44
  • Edited 11:40am Mar 16, 2007 11:37am | Edited 11:40am
Quoting Bemac
I suspect when you 'gladly share your winning method' you will soon be rolling in love as you watch someone else succeed because he is now rolling in dough. Why, he placed a copyright on it and, now it costs $x for 'your winning method'
Just something to consider in the future perhaps.
This pedantic retort predicates the very point I was making via example. In my experience, onerous prattle like this will go away if ignored.
  • #45
  • Mar 16, 2007 12:04pm Mar 16, 2007 12:04pm
When one has a trading system that is uniquely theirs in some aspect and back testing proves to them that it is profitable then one has created for themselves a psychological edge. One is confident in trading it.

Back testing is important in a system that is is/not yours for the exact same reason. If you do not have the confidence that it works 'For You' then you will move on to something else until u do find something that you are confident in.
In trading, there is no bullshit. You either make money or you don't.
  • #46
  • Mar 16, 2007 12:33pm Mar 16, 2007 12:33pm
Quoting FXTerminator
This is of course a myth, the forex market is so vast and so liquid it does not give a hoot how we are trading or what we are "up to". We are just small, small sardines swimming in the Pacific.
Yeah.

Quoting FXTerminator
For instance, classic technical analysis patterns like head and shoulders, trendlines, support/resistance, triangles, double top, double bottom, etc... have been around for more than 50 years, million of people are acting on these patterns every single day, on the forex, futures and stock market and yet, they are still working just fine like if they have just been discovered yesterday.
I'm not so sure about this. The forex is notorious for being the hardest market to trade the classic technical patterns. There's alway so many false signals. Take a simple horizontal range breakout, I can't remember the last time I've seen one with a clean breakout. It seems to usually pop out just a bit, then pull back inside the range, do it again, maybe this time on the opposite side; another false break. It continues like this until what you originally thought was the trading range has become a messy forest of choppy bars...you have to redefine your range high and low.

And the reason why is because there are so many technical traders trying these pattern...that mixed with the fact that the forex is essentially a zero-sum game. We are all buying form someone who is selling, and we are all selling to someone who wants to buy. If you are going to win money, you must steal it from another trader.

If we all got together and decided to buy on the break of a new high in a particular setup; let's say we all place a limit buy order just above the old high....it wouldn't work. First of all price would never even get to the new high because there are no buyers pushing it up...someone would have to use a market order so drive the price higher. Even if it did some how move up to our buy orders, the orders would not get filled because there are no sellers who want to sell it to us.

I hope I'm wrong. If there is a simple pattern that works consistently, please let me know. I think the classic patterns can be traded, but you need to put a twist on it: e.g. maybe wait for a couple false breaks before you try for the real thing.
  • #47
  • Mar 16, 2007 12:59pm Mar 16, 2007 12:59pm
Funny how the people arguing the case for sharing a system have been around for less then 6 months and have nothing of their own to give away.

FYI, even $3tril per day in total forex volume has to be divided across the 130 or so tradable pairs, across the entire daily range of each pair, and on both sides of each market. If all pairs moved their entire daily range, without a single retracement (which never happens), that would put average per pip open interest at less than $120mio. There are many more people who can move price then you would think.

Finally, if head and shoulders/box/triangle patterns are so reliable, then you should trade them. Hell, if you’re so convinced then you should quit your job, mortgage your house, max your credit cards, and throw it all behind those super reliable patterns. I'll be sure to drop my change in your cup on the way out of the restaurant.
  • #48
  • Edited 1:36pm Mar 16, 2007 1:23pm | Edited 1:36pm
Quoting mathematician
The forex is notorious for being the hardest market to trade the classic technical patterns. There's alway so many false signals. Take a simple horizontal range breakout, I can't remember the last time I've seen one with a clean breakout.
Your range breakout technique does not seem to work because you THINK you are looking at a horizontal range breakout. After all, how do you scientifically define a horizontal range breakout?

I could hand out the same exact forex daily chart to 100 people and they would all draw different trendlines, resistance/support lines and so forth.

Our own inability (charting is an art and requires a LOT of experience) to draw the "correct" trading range/pattern all the time is not a proof that these patterns are not reliable.

WE (the traders) are the "weakest link", there is nothing wrong with those patterns.

Quoting mathematician
And the reason why is because there are so many technical traders trying these pattern...
Yes, we hang on to that MYTH because we need to blame someone or something for our own failure. http://www.forexfactory.com/images/icons/icon6.gif

Quoting mathematician
that mixed with the fact that the forex is essentially a zero-sum game. We are all buying form someone who is selling, and we are all selling to someone who wants to buy. If you are going to win money, you must steal it from another trader.
No Sir, when you make 100 pips you don't necessarily "steal" it from someone else, you forget that big "500 Fortune" companies like BMW, Boeing or Coca Cola use the Forex to control (hedge) their exposure to currency fluctuations, they don't give a damn who is winning "their" money, they are not in the currency speculation business, they just want to hedge their positions, even though some big companies DO sometimes speculate on the FX market.

Quoting mathematician
I hope I'm wrong. If there is a simple pattern that works consistently, please let me know.
Those patterns are all STILL working consistently since MacGee and Edwards, you just need to practice every day.

In fact if you are not up until 4 AM in the morning looking at the darn charts and drawing trendlines and backtesting systems until your face turns blue then you are not really a trader
  • #49
  • Mar 16, 2007 2:31pm Mar 16, 2007 2:31pm
[quote=FXTerminator;246337]

I could hand out the same exact forex daily chart to 100 people and they would all draw different trendlines, resistance/support lines and so forth.

Our own inability (charting is an art and requires a LOT of experience) to draw the "correct" trading range/pattern all the time is not a proof that these patterns are not reliable.]


I agree with most of what you said except I think the above in bold is a contradiction. It is only an art if one does not use hard and fast rules to plot their various lines. Tom Demark is excellent on this.
  • #50
  • Mar 16, 2007 2:40pm Mar 16, 2007 2:40pm
FXTerminator:

I define a horizontal range as a sequence of prices where there are several lows that all come down to the same price (within a few pips) and also there are several highs that all go up to the same price. So there are several collinear highs and several collinear lows; if you were to draw a line connecting these highs and lows, they would be horizontal lines.

It's pretty hard to mess that one up. That's the kind of pattern where 100 out of 100 chartists would all draw the same lines.

If this pattern does in fact work consistently and if it's not going to hurt its effectiveness by telling me, then please tell me: what currency pairs and what time frames do you use to trade this pattern, where do you place your stop, in the middle of the range? About what percentage of the setups are successful and do not give a false breakout?

And no, I don't blame others for my failed trades. I know I am the only one to blame; I made a bad trade. I thought that simple S&R breakouts were going to be reliable, at least reliable enough when reward/risk, i.e. TP/SL, is taken into account. It wasn't working for me so I blamed myself: I picked a bad strategy. But you're telling me it that it does work. That's good news! So let's hear exactly how it should be done. Teach me. I'm always open to learning.

-mathematician
  • #51
  • Mar 16, 2007 2:42pm Mar 16, 2007 2:42pm
[quote=SunTrader;246386]
Quoting FXTerminator

I could hand out the same exact forex daily chart to 100 people and they would all draw different trendlines, resistance/support lines and so forth.

Our own inability (charting is an art and requires a LOT of experience) to draw the "correct" trading range/pattern all the time is not a proof that these patterns are not reliable.]


I agree with most of what you said except I think the above in bold is a contradiction. It is only an art if one does not use hard and fast rules to plot their various lines. Tom Demark is excellent on this.
Yes you are right, Demark has come up with some "scientific" ways to draw trendlines, like fractals bars, each with a different "power". But hey, this is only an approximation.

These S/R lines are the key to the vault because these lines are definitively the ULTIMATE non-random entry/exit points!!

Imagine the spectacular profits If only we could draw them with 100% accuracy all the time.
  • #52
  • Mar 16, 2007 3:04pm Mar 16, 2007 3:04pm
Quoting mathematician
FXTerminator:

I define a horizontal range as a sequence of prices where there are several lows that all come down to the same price (within a few pips)
Stop right there, what do you call "a few pips"?? 1 pip, 10 pips, 25 pips?
See, there is no scientific number with patterns, so you cannot say that pattern X is generating a lot of "false" breakouts simply because there is no way to define scientifically the pattern in the first place.

Quoting mathematician
If this pattern does in fact work consistently and if it's not going to hurt its effectiveness by telling me, then please tell me
I cannot, for the simple reason that almost every known pattern is successful but AFTER the facts, when the pattern is completed and the outcome known.

We can only guess that the pattern we are looking at is indeed pattern X. Then we use the proper stops and make our bet. Win or lose it does not matter, if we keep betting on those patterns that way we will win in the long run, because we are simply betting that people will react the same way when confronted with the same situation, that's why these patterns had stood the test of time.

But again charting is an art, like playing the piano, it takes a lot of practice and unfortunately I cannot practice for you, or for any trader here at FF.
  • #53
  • Mar 16, 2007 3:31pm Mar 16, 2007 3:31pm
Quoting Darkstar
Funny how the people arguing the case for sharing a system


Maybe I misread the posts here, including my own, but I understood that the argument was that IF someone wanted to share her system, THEN she shouldn't be afraid to do so. No opinion pro or con sharing was expressed.

Quoting Darkstar
have been around for less then 6 months
Sure. Igrok has "been around" for 4 months only, therefore he has no right of speech.
Quoting Darkstar
and have nothing of their own to give away.
Quoting Darkstar
I understand that you checked every single post ever made by participants of this thread, as well as checked their future intentions, and summarized your results in the statement above.

This hasn't been a nice post, Darkstar. Everybody is trying to contribute to the best of her or his ability, expertise, and experience.
sans peur et sans reproche
  • #54
  • Mar 16, 2007 4:04pm Mar 16, 2007 4:04pm
[quote=FXTerminator;246398]
Quoting SunTrader

Yes you are right, Demark has come up with some "scientific" ways to draw trendlines, like fractals bars, each with a different "power". But hey, this is only an approximation.

These S/R lines are the key to the vault because these lines are definitively the ULTIMATE non-random entry/exit points!!

Imagine the spectacular profits If only we could draw them with 100% accuracy all the time.
Of course no one can draw lines that are 100% accurate as far as profit/loss is concerned. But if you have a mechinical way of drawing them then there is But io ambiguity as to proper interpretation. For instance Tom DeMark starts from the right hand side of the chart (most recent data therefore the most important). Most others do it the traditional way of left to right. If you have any of TD's three books (I have all 3 as well some of his software, not bootleg versions) then you know the other rules. There are probably others that have mechanical rules based trendlines too that I haven't heard of.

If rules are applied, IMO it is no longer an art but a science.
  • #55
  • Edited 5:07pm Mar 16, 2007 4:05pm | Edited 5:07pm
Quoting Darkstar
Funny how the people arguing the case for sharing a system have been around for less then 6 months and have nothing of their own to give away.
So if I talk about beers in a forum for example that means I need to have a truckload of Heineken to give away first?

Quoting Darkstar
FYI, even $3tril per day in total forex volume has to be divided across the 130 or so tradable pairs, across the entire daily range of each pair, and on both sides of each market.

Not quite, over 90% of the forex volume is done on the 7 majors against the dollar. 1/3 is done on the EUR/USD alone.

Quoting Darkstar
There are many more people who can move price then you would think.
Then next time move the forex market by 75 pips in your favor and you will do just fine :

Quoting Darkstar
Finally, if head and shoulders/box/triangle patterns are so reliable, then you should trade them.


I already do

Quoting Darkstar
Hell, if you’re so convinced then you should quit your job, mortgage your house, max your credit cards, and throw it all behind those super reliable patterns.


I am working on it.

Quoting Darkstar
I'll be sure to drop my change in your cup on the way out of the restaurant.
Who said you were welcome in MY restaurant in the first place, I already bought that restaurant with my forex profits!

Thanks for the good jokes Darkstar, really
  • #56
  • Mar 16, 2007 4:07pm Mar 16, 2007 4:07pm
Quoting FXTerminator
We can only guess that the pattern we are looking at is indeed pattern X. Then we use the proper stops and make our bet. Win or lose it does not matter, if we keep betting on those patterns that way we will win in the long run, because we are simply betting that people will react the same way when confronted with the same situation, that's why these patterns had stood the test of time.
Ok, I see what you're saying now...We must simply assume we have picked the right pattern when there is only part of it showing. Trade it with stops and good money management and hope that it develops into the pattern we were betting on. Thanks!
  • #57
  • Mar 16, 2007 4:13pm Mar 16, 2007 4:13pm
Quoting mathematician
Ok, I see what you're saying now...We must simply assume we have picked the right pattern when there is only part of it showing. Trade it with stops and good money management and hope that it develops into the pattern we were betting on. Thanks!
Well said and right on the money Mathematician!
  • #58
  • Mar 16, 2007 4:20pm Mar 16, 2007 4:20pm
Quoting FXTerminator
Well said and right on the money Mathematician!
I read Igrok's book and it sounds like it is in-line with what you were saying: One should have a set of templates used to identify high probability setups. These are based on chart patterns but they are not so rigidly defined, there is some room left so it can be adapted to the current market situation.

Stops should be placed where the probability of a reversal is high...in fact it is often a good idea to actually reverse and take the opposite position if your stop is hit.

I am starting to come around to these ideas. One question: On what time frame do you typically look for patterns? Hourly chart, daily?
  • #59
  • Edited 5:34pm Mar 16, 2007 4:59pm | Edited 5:34pm
Quoting mathematician
in fact it is often a good idea to actually reverse and take the opposite position if your stop is hit.
Careful again Mathematician, this idea sounds extremely "logical" but you NEED to backtest it for your own particular system.

If you stop is hit it does not necessarily mean that now you need to change horses and bet the other way. I backtested a lot of systems and this idea does not seem to make money consistently (at least not all the time), in fact sometimes you just lose a second time when you do a stop and reverse.

So always test ANY trading idea, no matter how wonderful it sounds. Like I said before, most winning trading systems are counter-intuitive, while the seemingly "sound" trading systems are the biggest losers.


Quoting mathematician
I am starting to come around to these ideas. One question: On what time frame do you typically look for patterns? Hourly chart, daily?
I trade the 15 min, I like to see how prices react and behave when they touch or approach a support/resistance line between 0 and 5 AM EST: are they going to breakout or do we have a "price rejection"?

Then according to the price behavior I pull the trigger, with the appropriate stop and/or a stop/reverse as well, like I did today on the GBP/USD: I was stopped at $1.9430 for a 30 pip loss (I was short) then I reversed and bought the cable and I got my money back.

Of course it takes years of studying thousand of intra-day charts (I printed out 2000 charts already and my room is full of them) to be able to do that.

So print those charts out, study them, spend enough time with them and they will reveal their secrets to you...

In short be like Edison, at the end he did succeed and invented a working light bulb, but ONLY after he had run out of things that do NOT work.
  • #60
  • Mar 16, 2007 6:16pm Mar 16, 2007 6:16pm
everyone talks about how much money goes through forex a day. I thought we should break it down.

2 trillion dollars a day. thats $2,000,000,000,000

1 lot is $100,000

so 20,000,000 or 20 million lots a day in forex

nasdaq traded 2,143 million stocks today

so yes, lots of money is passing through, but there arent that many transactions.

Of those 20m lots traded its pretty easy to guess that institutions and goverments are placing 98% of those orders. Since most of them are profitable you can assume speculators are not taking money from them and they are not taking it from each other, they are taking it from speculators. This brings the amount that speculators trade a day about 500,000 lots.

so the question is, does publishing your system in effect destroy it. you betcha!

notice that turtle trades dont work anymore
you might notice that the reverse turtle trades dont work anymore.

there are 2 reasons for this.
1) you have to do the opposite of 90% of other people to make money
2) if money starts getting pulled out of other peoples systems they figure how to fix it. it might not be directly against your system but their optimization takes out your edge.

One big thing to point out too. after 100s of hours of testing its your system, you can share or not share. You are not a "bad" person for not sharing. You are a "bad" person for expecting hand outs.

as for the head and shoulders being "still good". they are great, after they happen. if you could predict them time, dont you think everyone would be rich now?
1 trader viewing now