The Ultimate Fallacy in Technical Analysis | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Technical Analysis Fallacy 63,567 replies

"Technical Analysis Fallacy" thread people, I need your help 54 replies

Technical Analysis Fallacy Redux 23 replies

Statistical analysis fallacy 33 replies

Attachments: The Ultimate Fallacy in Technical Analysis
Exit Attachments

The Ultimate Fallacy in Technical Analysis

  • #61
  • Feb 15, 2008 8:33am Feb 15, 2008 8:33am
Quoting Micro-MiniMe
If an indicator is based upon a mathematical formula such as a fib, and historical charts show price reacting at or at least very close to that level a very high percentage of the time, 75% of the time during the last year for example, would you concider this indicator usefull as part of a system?
Not if it did not make sense logically. For example, we all know that a simple MA cross is not a reliable signal, but I can curve fit it over the last year to a setting which could ultimately outperform every major fund out there. But next year the same trading rules might deplete over half my account using the same set of parameters. Same goes for price action and fib lines ect.

Quote
What I'm getting at is this. Just because a recent high or low has occured only indicates that either a large number of traders OR a few traders with large positions caused price to reverse. Is it logical to read the minds and motives of other traders without a means of knowing for sure what their intensions are and what actually caused the high or low to happen, or are we fooling ourselves into thinking we can read such cause and effect? Is not history repeating itself, as in the example with a fib, just as logical or even more so in this case?

When price moves up there are more buyers then sellers, demand is higher then supply.. end of story.. There is a reason why people buy at particular levels, but why go into that debate, dont fight price movements, just go with the flow. Trying to combat price movements can only lead to the destruction of one's account.

To avoid getting sucked into the non-warrented price movements, trade only larger time frames.
  • #62
  • Edited 8:55am Feb 15, 2008 8:44am | Edited 8:55am
Quoting superdezign
If you had done the research to prove to yourself that waiting for the bar to close before entering a trade not only gave you an added edge, but made sense logically to trade in such a way then you probably would have stuck to your plan this week and made a lot more money

its important that I emphasize, not only should your plan give you an edge but make logical sense as well.
My post was not very clear. Had I not stuck to my plan, I would have had a very good week. By waiting, I missed many good moves. On the trades I did enter according to plan, only one made a profit, just barely.

What you're saying above is that if somone has a statistically validated system, that they will make money every week if they follow it. I would have to disagree.
  • #63
  • Feb 15, 2008 8:48am Feb 15, 2008 8:48am
Quoting blueruby
You didn't read my post correctly. Had I not stuck to my plan, I would have had a very good week. By waiting, I missed many good moves. On the trades I did enter according to plan, 1 made a small profit.
my apologies, I skim more than I read and therefore miss a lot of important details.

ok, well in that case you have to ask yourself if actually waiting for the bar to close makes sense. Why do you wait for the bar to close?

im sure the only answer to that is "for added confirmation" well how does that add confirmation? does it at all really?
  • #64
  • Feb 15, 2008 9:08am Feb 15, 2008 9:08am
Quoting superdezign
my apologies, I skim more than I read and therefore miss a lot of important details.

ok, well in that case you have to ask yourself if actually waiting for the bar to close makes sense. Why do you wait for the bar to close?

im sure the only answer to that is "for added confirmation" well how does that add confirmation? does it at all really?
My post wasn't very clear. When I read it, I hardly know what I meant.

As an example, let's say the market settles down to a support level, then we get a long red bar that penetrates support - if it closes below support, this is a bearish signal, and if I had one more reason to, I would go short. If it retraces and forms a hammer with a long wick that penetrated support but did not hold, I would be looking for other signs of a reversal. After another candle or two closed, I might go long.

It does hurt the risk:reward, but improves probability over jumping in when price crosses support. I'm looking at possibly scaling in, take a small position when it crosses, but be just as aggressive at exiting quickly; after confirming candles, take the rest with a wider stop. I'm not sure I like that approach any better, but it would have worked this week.

This week, the candle pierces, then keeps right on going, quickly. When the candle closes, the opportunity is gone. People who enter on a stop had a great week. Another of my rules is never to think of "what could have been", "what I should have done", and especially "how much I could have made". These thoughts only lead to losses.

I do analyze losing trades, but from a viewpoint of improvement, not regret.
  • #65
  • Feb 15, 2008 9:21am Feb 15, 2008 9:21am
Quoting blueruby
My post wasn't very clear. When I read it, I hardly know what I meant.

As an example, let's say the market settles down to a support level, then we get a long red bar that penetrates support - if it closes below support, this is a bearish signal, and if I had one more reason to, I would go short. If it retraces and forms a hammer with a long wick that penetrated support but did not hold, I would be looking for other signs of a reversal. After another candle or two closed, I might go long.

It does hurt the risk:reward, but improves probability over jumping in when price crosses support. I'm looking at possibly scaling in, take a small position when it crosses, but be just as aggressive at exiting quickly; after confirming candles, take the rest with a wider stop. I'm not sure I like that approach any better, but it would have worked this week.

This week, the candle pierces, then keeps right on going, quickly. When the candle closes, the opportunity is gone. People who enter on a stop had a great week. Another of my rules is never to think of "what could have been", "what I should have done", and especially "how much I could have made". These thoughts only lead to losses.

I do analyze losing trades, but from a viewpoint of improvement, not regret.
arent closing prices just a figment of someone's clock? especially in forex..

if price goes below a support level doesnt that violate the support whether it closes below it or not?
  • #66
  • Feb 15, 2008 9:47am Feb 15, 2008 9:47am
Quoting superdezign
arent closing prices just a figment of someone's clock? especially in forex..
It depends. Daily charts can vary, but hourly charts seldom do - the hour begins and ends at the same time everywhere on the planet. Different data providers can have different charts, but you have to use something.

Quote
if price goes below a support level doesnt that violate the support whether it closes below it or not?

It depends on what you mean by "violate". I call it "piercing", or "breaching". If it closes below, I call that "breaking". I wait for a close below, I want the market to tell me what it's doing, a little more decisively. If I'm not comfortable, I'll wait for a candle to close below that candle's low before initiating a trade. Yes, I watch some good trades go by. There will be another.

Yes, I know, a 30M can close below, then you get "twin towers" or "bullish engulfing" and it goes back above support and closes, whereas a 1H candle will make a hammer. That's why higher timeframes can be better for some traders. They often tell a very different story than lower tf's. It's usually good to consult several t/f's.

T/F depends on YOUR clock. Some people trade the 5 minute, I can't handle it.
  • #67
  • Edited 2:31pm Feb 15, 2008 2:18pm | Edited 2:31pm
Quoting superdezign
Not if it did not make sense logically. For example, we all know that a simple MA cross is not a reliable signal, but I can curve fit it over the last year to a setting which could ultimately outperform every major fund out there. But next year the same trading rules might deplete over half my account using the same set of parameters. Same goes for price action and fib lines ect.



When price moves up there are more buyers then sellers, demand is higher then supply.. end of story.. There is a reason why people buy at particular levels, but why go into that debate, dont fight price movements, just go with the flow. Trying to combat price movements can only lead to the destruction of one's account.

To avoid getting sucked into the non-warrented price movements, trade only larger time frames.
Hi superdezign,

May I introduce to defer in your statement.
"That prices moves up when there are more buyer than sellers, and it is just that."

Generally, this is true, in markets, that are 100% efficient.
ie Structured markets like futures exchanges.

For OTC markets, where there exist a counterparty, ie market makers or stockist. What really happens is that the market makers will absorb the load from the customers and try to seek parties within the circle to help shoulder the exposure. In reality, when the Mmakers are taken/given on large, the spread WIDENs for a little (commonly known as shading) and the positions are to be attempted to be distributed amongst the circle of Mmakers. This activity can last for as long as 30mins before any real movement can be detected.

In reality, if there existed any chance of reprise, the markets is swing the other way around to attract for offloading of exposures. Thereby counter sweeping to offload.

I hope that you can understand the true machanics of the different structures.
In my experience, I had held position exposures so long as the next trading zone for offloading.All the time keeping market in a tight range. (scutter band)

regards
  • #68
  • Feb 23, 2008 5:27pm Feb 23, 2008 5:27pm
Quoting blueruby
My post was not very clear. Had I not stuck to my plan, I would have had a very good week. By waiting, I missed many good moves. On the trades I did enter according to plan, only one made a profit, just barely.

What you're saying above is that if somone has a statistically validated system, that they will make money every week if they follow it. I would have to disagree.
Or maybe I just forgot Axiom 3.

http://www.forexfactory.com/showpost...0&postcount=22
  • #69
  • Feb 24, 2008 12:16pm Feb 24, 2008 12:16pm
Quoting blueruby
Or maybe I just forgot Axiom 3.

http://www.forexfactory.com/showpost...0&postcount=22
I think that statement should have some fine print below it..

markets do evolve. however, there are certain underlying principals that are not impacted by changing conditions
  • #70
  • Feb 24, 2008 12:46pm Feb 24, 2008 12:46pm
Quoting superdezign
I think that statement should have some fine print below it..

markets do evolve. however, there are certain underlying principals that are not impacted by changing conditions
Those principles would be axioms, so I'd be glad to see them added to that thread. Just remember, to be an axiom, it has to be self-evident (needs no proof), and always true. We only came up with 7.
  • #71
  • Feb 25, 2008 8:31am Feb 25, 2008 8:31am
Quoting blueruby
Those principles would be axioms, so I'd be glad to see them added to that thread. Just remember, to be an axiom, it has to be self-evident (needs no proof), and always true. We only came up with 7.
Well, an axiom is very hard to come by in an environment that is left up to so much interpretation and at the same time is very ambiguous. Anytime someone refers to something being an axiom it should be taken with a very large grain of salt.

I have spent a lot of time (over two years of research) developing what I believe to be the only actual truths that mingle in the markets. So, that being said, I would rather not just spill all my hardwork over an internet forum. However, I will say this.. dont complicate things.. life is already hard enough
  • #72
  • Mar 3, 2008 6:54pm Mar 3, 2008 6:54pm
just frreaking trade, goddamit.
  • #73
  • Mar 3, 2008 7:10pm Mar 3, 2008 7:10pm
Quoting jest1081
just frreaking trade, goddamit.
LOL

yea.. ok..
  • #74
  • Mar 4, 2008 6:24am Mar 4, 2008 6:24am
Quoting superdezign
Well, an axiom is very hard to come by in an environment that is left up to so much interpretation and at the same time is very ambiguous. Anytime someone refers to something being an axiom it should be taken with a very large grain of salt.

I have spent a lot of time (over two years of research) developing what I believe to be the only actual truths that mingle in the markets. So, that being said, I would rather not just spill all my hardwork over an internet forum. However, I will say this.. dont complicate things.. life is already hard enough
Axioms simplify things to the most basic level of truth that can be attained.

They have to be general (not ambiguous) in order to always be true.

Take them with a grain of whatever you want, but if you can "bust" any of them, feel free.

  • #75
  • Mar 5, 2008 5:44am Mar 5, 2008 5:44am
I wholeheartedly agree with you, superdezign. Trading without understanding the logic behind each decision you make is financial suicide. Using indicators (fundamental or technical) as decision aids should be undertaken with great discretion.

That said, statistical analysis (the dead horse you'll hear me and a few others beat repeatedly on this forum) can be a powerful tool in financial markets, especially when combined with human intellect and psychology. To illustrate my point, I want to tell you a story I read recently:

As many of you may know, Netflix has a great system that suggests new movies that you might like based upon the ratings you've given to movies you've seen already. Many of their suggestions are good, but some are way off. To improve this service, Netflix has been running a contest since October 2006 that entices statisticians and programmers to try their hand at the task of automated movie-rating prediction. The goal: achieve an accuracy 10% better than Netflix's current algorithm. The winner of this contest gets $1,000,000 in cash.

The first groups to flock to this competition were the usual fare: academics. These are guys who spend their whole careers building and tweaking statistical algorithms to tease out correlations in data. They've so far done quite well: after over a year of work, they've achieved an 8.82% increase in accuracy. They've got their eyes on the prize and they're slowly creeping towards it.

But if you look at the Netflix prize leaderboard, you'll see something that should be inspiring to individual traders like most of us: "Just a guy in a garage". It turns out that this Joe Average is actually a 48-year-old retired management consultant with an undergraduate degree in psychology and a master's in operations research. Instead of approaching the problem of prediction from a purely mathematical perspective, he's picking it apart from its first principles. Like us, he's saying, "These are not all just numbers. These are decisions made, in some form or another, by people." That's how he's reached number 8 out of 26,939 teams in the competition. He's applying human psychology and logic to make hypotheses, and then using the power of modern computational tools to (in)validate those educated guesses.

If anything, I think that's the point I'm trying to drive home here. None of us have the statistical prowess that the guys at Bell Labs (AT&T) or MIT do, but we have an understanding of market behavior gleaned from years of experience. Apart from this experience, we also understand what I think of as the physics of the markets - what drives price movement (e.g., order flow). From this knowledge, we can posit intelligent hypotheses about market behavior, and then backtest those on a wealth of historical data.

Once we find a pattern or formula (rooted in logic) that seems to work on a regular basis, we can program that into an indicator so that we can more quickly analyze new situations or unfamiliar markets, increasing our decision accuracy and speed. We might even be able to create programs that can recognize trade opportunities using combinations of these indicators, thus enabling diversification across many markets without additional effort. This is the proper root and purpose of technical analysis - the recognition of common patterns for efficient prediction of future market movement, whether through statistical formulae or visual inspection.

In 29 words: Experience and understanding lead to hypotheses, testing leads to theories, theories lead to good trades, and good trades lead to profit. That's how we, the mathematical underdogs, can win.
  • #76
  • Nov 30, 2008 9:56am Nov 30, 2008 9:56am
Quoting Leugimp
I dont agree. Knowing what makes up an indicator and what "limits" this makeup creates allows you to use it in ways that would be rather difficult to do with just price. One cannot precisely measure speed and acceleration in a visual fashion just by looking at time and displacement. If that is important to you I would suggest learning the math that would allow you to measure such properties. Not to say that you cant trade without it, but lets not evaluate its usefullness by the words of those that find them useless.

What makes trendlines any...
Leuqimp:
You're on the right track. There all kinds of indicators, if you want to look
for them. First of all....Price is a function of time...P=f(t). Your mention
of the sine angle for the slope is in the Y=mx+b formula. But you have to
go just a bit further. The run of the market goes into expondental and
logarithm, switching from one to the other, so you find yourself using the
slope of the secant. There's anothert little thing you should know. The only
truth is the "tick"...all the rest is history. You should automatically be able
to read the "chart" and use the indicators as you implied. I've been study-
ing the market for the past 45 years...and I'll be damn if I will ever get to
know it. I made it in commodities, but that was years ago. This forex market
is another beast and what worked for me in commodities is not the same
for the forex market. For one thing....you don't have "holes" in the line
like those in stocks and commodities. That is...the forex is a 24 hour thing,
while the stocks and commodities are in a few hours. Thus the gaps.
Regardless....money management should be on the top of your list if you
want to eventually walk to the bank. Also, the odds are against you...no
matter how good you are in trading if you don't have the proper capital
and know how to use it. You have a better chance trading the forex than
the other markets. Look....we've all been brain-washed by all the past
gurus. My advice to a newbie is to "STAY OUT OF THE MARKET....you
probably don't belong in it. It's tough as hell. I think you're better off
in going down to the local pub and spend a few shekels on a good glass
of beer and enjoy it. And you might even meet some beautiful looking
chick in the process. And after she gets through with you...you'll
probably wish the hell you kept trading by going 'short' GBP/USD. But
you...Leugimp, seem to have the right size gloves...and it fits.
  • #77
  • Dec 1, 2008 3:25pm Dec 1, 2008 3:25pm
The whole idea of trading is creating a good strategy. You can use your mix of indicator and create your own expert advisor…or you can try one from somebody else. But the point is to be comfortable with your system and your rules. I’m not talking about some psychological bullshit; I’m talking about confident and following rules. I have tried some strategies and even when they have been backtested for some great traders, any of them have worked for me. The successes of a strategy depend on the confident you get from it when taking a decision. You must be sure of how your strategy works in different situations. In other words, you must understand your strategy and be able to follow your rules under any condition. And that’s the big deal with strategies; they can be profitable and successful but if you are not able to understand the picture they are showing you will lose.
  • #78
  • Dec 1, 2008 3:56pm Dec 1, 2008 3:56pm
I think the rules are:

Step #1:
1. Include two indicators, see if the first green bar in the positive side aligns.
2. Make a happy face and think you're a genious. It seems to work but it needs a little tweaking.
3. Include yet another indicator, this time some curve, and see if some of the "bad trades" are filtered by it.
4. Change parameters to the first indicator.
5. For some reason, add stochastics.
6. Open a thread saying that you've been trading this system for some months now, and you're positive it's 70%+ accurate.
7. See your thread go up and have a lot of followers.
8. Dream about your new Lamborghini.
9. Begin to see that some of your followers are just "Thanks for sharing, will test next week. How do I add the mq4 to the indicators folder?"
10. See your system chopping you and everybody else once the market finishes the trend in which you invented your indicator.

Step #2
1. Stop trading and think about you going back to your normal life, with your hated job.
2. Feel the frustration.
3. Wait a few days until your mouse suddenly opens MT4 again.
4. Go back to Step #1 once more, then go to Step #3

Step #3
1. Take your old book about Technical Analysis and wonder why so many people make money.
2. Go through some pages. See support & resistance.
3. See the charts.
Imagine... all this time and you had the solution in front of you since day 1.

Step #4
Trade without indicators.

------------------------------------------

Seriously, I think that most traders would make a lot more money just by sticking to babypips.com and Forex 101.

However (and this is the funny (maybe not so funny for me) part), people like William Eckhardt, and thousands of traders, seem to do very well without PA, just by systems and indicators.
In fact, I have some days trading with a new indicator and I've been over 30 consistently.

Could it be a Step #5, go back to Step #1?

The answer seems dangerous, but I think that if you've earned some experience in the process, what may seem a "going in circles" to the trader who sees it from a point above, may seem a "going up in spiral just like a parking silo" to the trader who sees it from a point in the side.
I guess you can clearly see if you're going up or going in circles. Check your bank account every time you go through Step #1.
  • #79
  • Dec 29, 2008 8:26am Dec 29, 2008 8:26am
Quoting Quantum4x
I think the rules are:

Step #1:
1. Include two indicators, see if the first green bar in the positive side aligns.
2. Make a happy face and think you're a genious. It seems to work but it needs a little tweaking.
3. Include yet another indicator, this time some curve, and see if some of the "bad trades" are filtered by it.
4. Change parameters to the first indicator.
5. For some reason, add stochastics.
6. Open a thread saying that you've been trading this system for some months now, and you're positive it's 70%+ accurate.
7. See your thread...
This is fantastic...I love it.
Empty as water and soft as a facing wind mountain
  • #80
  • Dec 29, 2008 4:42pm Dec 29, 2008 4:42pm
Quoting superdezign
The ultimate fallacy in technical analysis is that people do not understand how their indicators work, and if you can not figure out the logic behind what the indicator is telling you, it should not be used.
That's why I don't bother paying any attention to the speedometer in my car...
nav++
0 traders viewing now