Topic: A Japanese Candlestick Analysis - page 2. (Read 4247 times)

Which leads to the conclusion: what technical analysis claims to achieve is pointless.




We can safely skip the question if it's an intentional deception or mystification, it is sufficient that it is a mystification, by its nature.


"Technical Analysis".
The whole way it is named and is presented by it proponents suggests a degree of precision and rationality which in reality isn't there.

The first thing when you start learning math is to prove the necessity of new terms and laws you introduce, and to prove that any construction you create actually delivers what you claim it to do. The first thing when you start learning engineering is to judge the accuracy and error level of your measurements. And you learn that it is your duty to prove that an claimed effect actually exists.

Note that the point is not that you can generate an effect, but the point is to prove that a proposed method is necessary to yield the prediction, understanding or effect.

It is possible to create a stunning construction out of mathematical methods, which might be even hard to grasp and to execute, while it has actually no relation to either the nature of the mathematical objects, or any relation to the observable parts of nature or society. Such construction might be a piece of art, it is not math, nor is it analysis, nor is it a precise method.

It might well be that all the stunning constructions, formulas and indicators of technical analysis have only weak or no relation to the real phenomena, and that any observed "benefits" from applying technical analysis are actually due to the experience of the traders and due to just statistical effects. This possibility has not been sufficiently addressed with due diligence.

Thus the verdict of a mystification and self-deception.


Maybe (I hope so) current and future research will give us reliable methods and a positive proof of the effectiveness of such methods. Maybe this verdict can be lifted at some point. But we aren't there and we need more critical thinking and questioning of our own actions.


No doubt any effective trader has heuristics and guidelines.

But if it is experience, then please call it experience.
If it is gut feeling then please label it as gut feeling.
If it is a craftsmanship, gained by training and apprenticeship, then please call it so, and be proud of it.


But don't label it with anything associated with exact methods, science and technology.

And stop that math and numbers bullshit for things which have an prediction error of +- 100% and only apply in 3 of 5 cases. ;-)

But the noodles are in the pot and they're BOILIN' ... BOILIN' I tell ya!
 

all those who are doing TA aren't good enough data analysts an coders, to actually devise such an algorithm. second, if someone has it, they would keep it secret and make $$$. so, there isn't even any incentive to publish this.

and more generally, just because it doesn't exist NOW, doesn't mean that it is impossible.

are you projecting? it's not common sense, it must be learned. and the idea of the relationship between derivatives and integrals (rate of change vs cumulative height) is also very important. in other words, you need to know bitcoin, AND math. not very many do. that's why it's lucrative.

the reason the below is completely unfeasible:


is because markets are anti-inductive. they correct for 'knowledge' of future price information. so you not only have to see the trend forming from regular trading, but anticipate the actions of all of the others who see the trend as well.


again, just because you don't understand something doesn't mean it's intentional mystification. i have no doubt that the large subscription-based analyses primarily function by herding the 'ignorant' masses but don't let those few give all of us a bad name. and it's not like other methods are equal in prediction reliability; you can talk until you're blue about news or fundamentals but these things are extremely difficult to quantize with regards to price information. i.e., how is buying because you believe BFL will follow through an informed trade?

Exactly my point.
Technical analysis isn't vodoo, it is a mystification.

It is just a bunch of common sense packaged in arbitrarily complex terminology and a layer of math, which serves the purpose to impress the uneducated.


Likely, Likely, Likely! less, maybe, but not if.

This isn't analysis, this isn't engineering.
It's guesses, gut feeling and a lot of experience.


If technical analysis would be something substantial, then you could create a constructive algorithm, which could be executed by a machine or a human without any experience, and generate precise or at least >80% reliable predictions (on statements which have substantial explicative value)

Technical analysis can't deliver to that promise. So you can skip all the allegedly high tech smoke and mirrors, and just apply a bit of common sense, to arrive at the same level of predictions and reliability.

If you want to actually understand candlestick charts I recommend going here: http://www.babypips.com/school/

27/01/2013 Daily candle was a Doji

The hanging man pattern only means that a sale is'nt met by a buy of similar level up to previous price. The sale leaves a hole with no bids, which is often the case after a rise in price.

It does not always prelude a drop in price as a buyer might be waiting for that to push price up. In larger markets, not Bitcoin, the "fast in, fast out" traders might sell when they see hanging men, and it becomes an indicator.

Right, there are certain regions that the indicators won't go or won't go for long.  When you know a move in a particular direction would take the indicator somewhere it doesn't like to be then that direction is less likely to occur.  The money is in herding, not stampeding.

technical analysis isn't voodoo. it isn't something you believe in or you don't. most indicators are price/volume transforms and their abstractions. if you understand the behavior of price and volume generally (no rallies on decreasing volume, support/resistance at price levels, and retracements) then you'll understand technical analysis. i'm tired of this lemming bullshit. if you just go "LOOK THE PRETTY PICTURES ARE GOING UP UP UP" then you're not doing technical analysis.

technical analysis could be a self fulfilling prophecy.
The more people believe in technical analysis and act accordingly, the more the patterns "predicted" by techical analysis start showing up in real world measurment values.

So there are two aspects to this issue.

• if you want to prove scientifically that there is some real effect, as opposed to just an assumed effect, then you'd need a way to prevent your test subject to know the rules to be proved -- to prevent behaviour according to an assumed effect. Some kind of a double blind test
• if you're looking just for some practical guidelines, the more people know and believe in technical analysis, the better. Congratulations, you've produced a nice flock of mentally conditioned sheep. 

i'm  trying to figure out at what point those losing money trying technical analysis against bitcoin simply conclude that it's different with bitcoin.

[/quote]

do you have evidence for that positive claim?

[/quote]

i'm  trying to figure out at what point those losing money trying technical analysis against bitcoin simply conclude that it's different with bitcoin.

stuff to read http://www.amazon.com/Steve-Nison/e/B001IGNZSE

26/01/2013 daily candle looks like a hanging man