Pages:
Author

Topic: arepo's easy TA (Read 3975 times)

sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 09:48:13 PM
#60
To complement my personal opinion, I think there might be some important insight hidden, and the idea to perform an systematic analysis of the primary market phenomena (as oposed to just explain everything by "fundamentals") has merit. But the specific way this has been done in TA might not be totally appropriate, maybe needs to be challenged and readjusted to get us there.

i can agree here, and am always willing to entertain serious challenges, although i do sometimes take breaks from the forums Tongue

thank you for the very humble response.

--arepo
hero member
Activity: 602
Merit: 500
April 26, 2013, 09:40:56 PM
#59
Arepo, to give you a quick answer to the matters tone and reputation --
I'd happily step down the emotional level of this controversy; also I acknowledge that it can be very difficult to carry out a serious controversy in this forum; I for my part are far from shouting statements "technical analysis is total bullshit", and "thats the greatest nonsense I've ever heard" -- but other people do, making it difficult to stick to a civilised level.

Also I want to stress the point, that I have no objections against presenting some TA related stuff as personal heuristics.

But there is a serious discrepancy between the generality of the claims, the serious technological appeal of the presentation of methods, the attitude of presenting an agreed on and well proven set of tools, and the weakness of the actual definitions, reasoning and conclusions and the lack of critical counter checking in pretty much every presentation of TA methods I've seen. A method using the mathematical and scientific toolset should be prepared to be measured against the respective standards.

Quote
the calibre of proof for which you are asking demands hundreds of hours of work, and if you think i'll be willing to do that just so you can shout me down and determine just why it isn't 'scientific' then you'd be dead wrong.

The question is not if you, Arepo should do that work.
Rather the interesting question is, if that can be done at all, and how -- and if there was any serious attempt towards such an underpinning.

In the end, the reason why we push any tasks of engineering and science is not primarily to be "right" or "wrong", but to advance the general understanding, to build on each others work and to shed some light in one's own premises.



To complement my personal opinion, I think there might be some important insight hidden, and the idea to perform an systematic analysis of the primary market phenomena (as oposed to just explain everything by "fundamentals") has merit. But the specific way this has been done in TA might not be totally appropriate, maybe needs to be challenged and readjusted to get us there.

PS: thanks for the detailed answer, I'll respond to some points more in detail later
hero member
Activity: 602
Merit: 500
April 26, 2013, 09:40:42 PM
#58
You clearly know nothing about TA at all, so for the love of god go learn something before you resume bashing it.
Rusky, what makes you think so?

Actually I know a lot more than just the wikipedia article.

Read here for a primer on the arguments you're presenting and why you're wrong.
This link leaves me a bit puzzled. The very paragraph you cite mentions my and several further objections. And it shows a surprisingly thin evidence for methods promoted so boldly by many. Other sources I've seen didn't give me much confidence either.

Actually I hoped some of the TA proponents could point at some reliable standard source to dissipate those doubts.

sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 08:54:17 PM
#57
Quote
(1) this can not be concluded from the pattern you see in the graph. You need to look at the actual orders, and watch the actual deals going on in order to determine (a) if there was actually a resistance, (b) the resistance was the same which couldn't overcome by the first upward oscillation and (c) there are no further resistances, so you could call it rightfully the critical resistance.

(2) You can only conclude this in hindsight. We agree that there was a fluctuation in volume, after the downward movement ended. But when the volume gets up again, the movement can indeed overcome several levels of resistance. Or it can move sidewards between the various resistance levels. Or a huge wall just gets pulled away and this prompts new market action.

(3) Here you claim a generality which objectively doesn't exist. Only in hindsight, when you deliberately pick a consolidation break between several downward legs as an example, you make it look coherent as if there was a uniform and general causal structure behind the price movement. But you omit the cases (a) where you could see a triangle pattern, but no bull trap and no 'last ditch effort', and no continued downward move followed, and (b) you omit the cases where in a similar constellation there was a fluctuation in volume but no clearly shaped visible pattern formed before onset of the next market action.

(4) to stress this point: here you give the impression that you can provide a reason behind the observable action. You claim to have a better explanation than just a "signal" derived from watching the chart, using notions like "resistance", "trend", "consolidation".

(5) the model of classical mechanics indeed shows the virtues of a rational model. It is based on well defined axioms. And you can undoubtedly name the conditions when it is applicable and when not (preservation of momentum). The entities and forces detailed by this model can be shown to exist at the time the observed action actually takes place. Plus, this physical process has indeed a common causal nexus. It is one process, a transport of energy and momentum

(6) but what you provide here as an "physical model" of price action is borderline ridiculous. (not to mention that you don't define your terms and that you don't specify the conditions when this model can not be applied). At the time, when the real market action takes place, there is no generic common reason behind the upwards movement and the downwards movement. Other than the fundamental situation in any market that some people want to buy and some want to sell. That some people want the price to go up and others want it to go down. You can not claim in general that the momentum of the upwards movement causes the momentum of the downward movement. (And if you insisted on this claim, I would demand a first class proof)

There might indeed be a set of typical mass psychological situations, like people being nervous after a crash and anxiously watching the further movement to get an indication. But the way you are presenting your analysis here is not helpful to pinpoint such a situation.
  • (a) the pattern you present is formulated way too much generic and especially omits the specifics of the market and the current situation
  • (b) the pattern can only be construed reliably in hindsight, when the further action is already revealed ("confirmed" as you like to put it)
  • (c) the pattern is not necessary to determine a specific mass psychological situation, e.g. a crash. The fact that the price did not turn round but continue into a second leg down can be determined directly from the data available when the "confirmation" of the pattern has happened.

that being said, i really do appreciate stuff like this -- like i mentioned earlier your criticisms have, at least sometimes, been constructive and i have kept you in mind to help maintain rigor in my work. my methods are becoming more empirical thanks to you  Wink

--arepo
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 08:22:45 PM
#56
Arepo, the way you behave makes you look like an dishonest manipulator.

thank you for your opinion -- know that many disagree with you. i'm not going to babble about asking for "proof" of this claim, but know that whenever you make statements like this, you're treading on a double-standard.

Price movements are stochastic, meaning, like in quantum physics, they are impossible to predict with 100% accuracy.
...
Fortunately, although price movements are stochastic, they do exhibit time-dependent autocorrelations. This just means that there are deterministic rules in play as well, whose influence is visible through self-similar patterns.

This is a bold statement. You claim that you can show deterministic rules at work below a seemingly random surface.

yes, again, because there is proof that they exist. why shouldn't they be measurable?

more evidence of the rigor of some methods of TA can be found in this thread, where the OP and i came to the same conclusion based upon wildly different techniques. that's called science.

--arepo

-===-

highly successful, straightforward, repeatable, demonstrable, swingtrading technique

also, here's a very simple ceiling-bounce/'doji'+volume hunting exercise i was doing yesterday while daytrading. very straightforward and effective method:

-===-



-===-

--arepo

-===-

prediction of a double bottom

Something like this?



still possible, though i assess unlikely from the following:
 
bears lose ammo with every bounce, that is, bounces on high volume are 'exhausting' to the trend.

also, the two bottoms are too symmetric. i haven't noted in this thread, but this pattern should look familiar (see action after first crash to $50). the bottoms of the bounces are almost equal, and their volumes are almost equal.

in other words -- for falsifiability -- if this bounce had significantly lower volume, or did not reach as significantly close to $120, then your model would have better chances.

try to use your bullish intuition, too -- flip the picture upside-down, does it look bearish to you? (double top)

only time will tell...

--arepo

edit: not saying we can't still go down from here, but i think the most bearish situation would be a period of consolidation around this critical resistance before another significant leg down.

-===-

prediction of movement down from $160s

update 6:

10-day 2-hourhourly scale



-===-

'moving support' here even more robust than the scaled chart i sent out in emails, showing the insane momentum of this last push. we're gonna start running out of space, soon, though. picking up some risk here.



follow-up for above prediction

update 6:

more trendline magic!

William's %R 10-DAY 1-HOUR scale

-===-



-===-

downward breakout associated with a correction to the uptrend, as projected from update 5. 50-line bounce is very bullish, and we seem to have found support at a lower line.

prediction of movement down from the $140s

update 11:

*charts are 1-month 6-hour scale*



-===-

right now, we are still in scenario black.

if the trendline breaks, scenario red could still be bullish, but carries a lot of reversal-risk (see high-volume green 'doji' on 6-hour scale)

and at least for now, it still seems like there is too much bullish momentum and not enough selling pressure for scenario blue (REVERSAL).




-===-

this is it. this is the big moment where we test the trend's support -- if we go any lower, the picture turns bearish. the bullish momentum is gone and the selling pressure is on, we need a big bounce.

-===-

compare to:

I guess this was C, as trend line under attack



$140 should hold... if we go below then trend is broken...

[from lucif's thread again]

--arepo

-===-

also, you were incredibly dismissive about ruski's earlier comments, which referred to his own techniques. i understand that you should be suspicious because any of these methods are prone to apophenia, among other issues, but refusing to believe it's worth jack shit until you read a full statistical analysis of bitcoin price data and prediction methods is not intellectually honest. the calibre of proof for which you are asking demands hundreds of hours of work, and if you think i'll be willing to do that just so you can shout me down and determine just why it isn't 'scientific' then you'd be dead wrong. the same goes for your breakdown of the OP of this thread. i referred to my private models multiple times, and you see my 'withholding' of the statistical justification as manipulation, when the point of me posting on these forums is not to publish a rigorous proof of why my methods are effective (which would be long, boring, and also cause me an information loss against my 'opponents' -- other traders), but rather to exhibit these methods. this does not mean i have not done the due diligance on my own, as i would not dare publish anything that i haven't determined is sound in my own work.

if this post still doesn't satisfy you, please do not break-down each and every example and point at the holes, etc. i'm am getting tired of this game. i'd like to save us both time -- if you still feel adamant, please, please, please supply me with specifically what would constitute evidence for my methods, as i have attempted three times already in this thread, and now many more, to satisfy that most important claim. i have nothing, at all, to hide, and i would appreciate a little bit of cool-down regarding the claims of dishonesty.

--arepo
full member
Activity: 350
Merit: 100
April 26, 2013, 07:11:14 PM
#55
Ichthyo I swear you would've been the one arguing passionately that the world was flat in the 1500's.

Read here for a primer on the arguments you're presenting and why you're wrong.

You clearly know nothing about TA at all, so for the love of god go learn something before you resume bashing it.
hero member
Activity: 602
Merit: 500
April 26, 2013, 06:54:37 PM
#54
How would I do that if I couldn't recognize...

Please proof that there is no other way finding out.
Please provide a complete account of all incidents, where your method gave you the right indications, and all the incidents were your method led you astray.

Unless you can do so, the conclusion is: your "method" doesn't exist.

this is garbage -- really.
...
so go home and buy a lottery ticket. and for the last time, if you have something to say like "TA is bunk" or "this thread is garbage", keep in mind that you're doing exactly what you claim i am -- making groundless assertions.


Arepo, the way you behave makes you look like an dishonest manipulator.

You constantly turn and distort other peoples arguments, you create huge claims, and when challenged for a real proof, you walk away and start further threads.

if there are no 'VERIFIABLE METHODS WHICH YIELD POSITIVE RESULTS AT A RATE BETTER THAN CHANCE" than how can anyone be a 'skilled' trader? what does that even mean?

you are bending and derailing the argument.

I demanded proof that a proposed method actually yields results.
I demanded proof that the terms and proceedings are actually necessary to arrive at certain conclusions.

And you: you say "if there are no verifiable methods..."

This is how a manipulator proceeds.

Price movements are stochastic, meaning, like in quantum physics, they are impossible to predict with 100% accuracy.
...
Fortunately, although price movements are stochastic, they do exhibit time-dependent autocorrelations. This just means that there are deterministic rules in play as well, whose influence is visible through self-similar patterns.

This is a bold statement. You claim that you can show deterministic rules at work below a seemingly random surface.

Yet in any case where your statements were challenged in the last weeks, you were not able to show an objective rule, law, or otherwise verifiable formula, method or even just empirical proof.

When pressed by critics, you always retract to statements like "it needs to be confirmed by later price movement". Or "this thread was not intended to be predictive"


I'll take an example right from this thread.
You claim to show a causal nexus, but actually you only show a mental projection in hindsight.

C tests the critical resistance.(1) It is the interesting 'last ditch effort' of the suppressed upwards correction.(2) It is referred to as a "breakout" both because it breaks out of the tightening range, but also because the candle has a suddenly large volume in stark contrast to the waning volume inside the formation. This fully confirms the pattern.(3) This candle generally(3) reaches only as high as the highest point on the wide back-end of the triangle(3) (in this picture it only reaches halfway), but is a nasty bulltrap* because it signals the failure of the price to overcome the resistance (white line near C) which the price was consolidating against. This is followed by the continuation of the trend, D.

*Why(4) does this occur? Why not an immediate downside breakout into the trend continuation (4)? The 'signal' explanation (above) is decent in terms of basic understanding, but is not complete. In my physics model(5), it is equivalent to momentum. In order to jump, one needs to push against the floor with a force equal and opposite(5). In a way, the price is 'bouncing off of a ceiling' as opposed to 'jumping', where the price pushes against the critical resistance to gain enough momentum to continue the downward motion(6).

(1) this can not be concluded from the pattern you see in the graph. You need to look at the actual orders, and watch the actual deals going on in order to determine (a) if there was actually a resistance, (b) the resistance was the same which couldn't overcome by the first upward oscillation and (c) there are no further resistances, so you could call it rightfully the critical resistance.

(2) You can only conclude this in hindsight. We agree that there was a fluctuation in volume, after the downward movement ended. But when the volume gets up again, the movement can indeed overcome several levels of resistance. Or it can move sidewards between the various resistance levels. Or a huge wall just gets pulled away and this prompts new market action.

(3) Here you claim a generality which objectively doesn't exist. Only in hindsight, when you deliberately pick a consolidation break between several downward legs as an example, you make it look coherent as if there was a uniform and general causal structure behind the price movement. But you omit the cases (a) where you could see a triangle pattern, but no bull trap and no 'last ditch effort', and no continued downward move followed, and (b) you omit the cases where in a similar constellation there was a fluctuation in volume but no clearly shaped visible pattern formed before onset of the next market action.

(4) to stress this point: here you give the impression that you can provide a reason behind the observable action. You claim to have a better explanation than just a "signal" derived from watching the chart, using notions like "resistance", "trend", "consolidation".

(5) the model of classical mechanics indeed shows the virtues of a rational model. It is based on well defined axioms. And you can undoubtedly name the conditions when it is applicable and when not (preservation of momentum). The entities and forces detailed by this model can be shown to exist at the time the observed action actually takes place. Plus, this physical process has indeed a common causal nexus. It is one process, a transport of energy and momentum

(6) but what you provide here as an "physical model" of price action is borderline ridiculous. (not to mention that you don't define your terms and that you don't specify the conditions when this model can not be applied). At the time, when the real market action takes place, there is no generic common reason behind the upwards movement and the downwards movement. Other than the fundamental situation in any market that some people want to buy and some want to sell. That some people want the price to go up and others want it to go down. You can not claim in general that the momentum of the upwards movement causes the momentum of the downward movement. (And if you insisted on this claim, I would demand a first class proof)

There might indeed be a set of typical mass psychological situations, like people being nervous after a crash and anxiously watching the further movement to get an indication. But the way you are presenting your analysis here is not helpful to pinpoint such a situation.
  • (a) the pattern you present is formulated way too much generic and especially omits the specifics of the market and the current situation
  • (b) the pattern can only be construed reliably in hindsight, when the further action is already revealed ("confirmed" as you like to put it)
  • (c) the pattern is not necessary to determine a specific mass psychological situation, e.g. a crash. The fact that the price did not turn round but continue into a second leg down can be determined directly from the data available when the "confirmation" of the pattern has happened.

sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 06:07:13 PM
#53
[snip]

You cannot know from TA when I will realize to start/stop PUMP/DUMP. I know how much money I have $1000, $10,000, $100,000, $1,000,000, $10,000,000, $100,000,000, $1,000,000,000 but TA does not know. I'll stop when MY TA will see PANIC.

it has nothing to do with how much money the players have. you clearly did not follow my line of thought. are you familiar with game theory, by any chance?

anyway, regardless, larger players will have less and less of an impact as time goes on, by many mechanisms.

--arepo
legendary
Activity: 1414
Merit: 1000
April 26, 2013, 05:51:54 PM
#52
It is not you or your TA who CAN change the price. Market is SMALL, single ENTITY can PUMP or DUMP the price where wish to have. And PANIC will HELP.

now this is an actual point of contention, not FUD, thank you.

my thinking about this goes as follows:

TA demonstrates trends, supports and resistances, and other properties of the price function that help determine which 'moves' are profitable. if someone wanted to dump their coins, they could have done so at the $120 resistance, when we were threatening to crash right through. why didn't they? probably because they think they can get better than $120 per coin -- that is, they valuate BTC as "oversold". this is exactly why i'm anticipating a leg up. the sellers have run out of coin, and this fact is observable in the data. many are on the sidelines, but the selling pressure has abated.

--arepo

You cannot know from TA when I will realize to start/stop PUMP/DUMP. I know how much money I have $1000, $10,000, $100,000, $1,000,000, $10,000,000, $100,000,000, $1,000,000,000 but TA does not know. I'll stop when MY TA will see PANIC.
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 05:41:30 PM
#51
It is not you or your TA who CAN change the price. Market is SMALL, single ENTITY can PUMP or DUMP the price where wish to have. And PANIC will HELP.

now this is an actual point of contention, not FUD, thank you.

my thinking about this goes as follows:

TA demonstrates trends, supports and resistances, and other properties of the price function that help determine which 'moves' are profitable. if someone wanted to dump their coins, they could have done so at the $120 resistance, when we were threatening to crash right through. why didn't they? probably because they think they can get better than $120 per coin -- that is, they valuate BTC as "oversold". this is exactly why i'm anticipating a leg up. the sellers have run out of coin, and this fact is observable in the data. many are on the sidelines, but the selling pressure has abated.

--arepo
legendary
Activity: 1414
Merit: 1000
April 26, 2013, 05:33:58 PM
#50
arepo TA are BS. SELL SELL SELL (we are going to single digit)

someone's got to sell at the bottom... pigs gonna pig

and gtfo of my thread with this FUD, i'm seriously tired of it. i haven't even made any bullish calls here....

that being said, discussion and presentation of analysis ITT,

thank you

--arepo

p.s. you might also get the pig award for posting this moments before we resume the uptrend -- by my measurements, within the next 2 hours (if we break out now), or in the next 24 hours (if we consolidate downwards one more time from this test of the $140 resistance ).

It is not you or your TA who CAN change the price. Market is SMALL, single ENTITY can PUMP or DUMP the price where wish to have. And PANIC will HELP.
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 05:30:38 PM
#49
This--
[snip]

Thoughts?

good observation!

there is one very important difference. bearish wedges exhibit higher lows and higher highs, resulting in a general upward slope.

this triangle has a level top resistance, and we're consolidating bullishly so far. some evidence for this in this post.

also, this is an old thread that got necromanced, i'd like to keep all current analysis in a single thread, namely "arepo's detailed... ". i'm going to quote you there, this is good analysis.

--arepo
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 05:25:11 PM
#48
arepo TA are BS. SELL SELL SELL (we are going to single digit)

someone's got to sell at the bottom... pigs gonna pig

and gtfo of my thread with this FUD, i'm seriously tired of it. i haven't even made any bullish calls here....

that being said, discussion and presentation of analysis ITT,

thank you

--arepo

p.s. you might also get the pig award for posting this moments before we resume the uptrend -- by my measurements, within the next 2 hours (if we break out now), or in the next 24 hours (if we consolidate downwards one more time from this test of the $140 resistance ).
hero member
Activity: 826
Merit: 508
April 26, 2013, 05:18:29 PM
#47
This--


...reminds me of these:



Thoughts?
legendary
Activity: 1414
Merit: 1000
April 26, 2013, 05:15:03 PM
#46
arepo TA are BS. SELL SELL SELL (we are going to single digit)

EDIT: ahhh you already sold ? It is your problem. :-)
Market is so small, than a lot of entities can move price where they wish.
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 04:33:32 PM
#45
edit: the second reversal candle looks more like an inverse 'shooting star', but the small body, large volume, and large range proportional to the body are all characteristics of a reversal candle.

in fact, there's a way to understand this more intuitively. 'reversal candles' are periods of time in which the prevailing pressure (buying or selling) changes direction. since there is a struggle, or 'candle battle' between bulls and bears, the body tends to form very small, as the open and close are pulled close to each other by alternately competing forces, while the range of the shadow, or wicks, of the candles is pulled outward by strong waves in each direction.
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 04:26:28 PM
#44
also, here's a very simple ceiling-bounce/'doji'+volume hunting exercise i was doing yesterday while daytrading. very straightforward and effective method:

-===-

[img]

-===-

--arepo

Sorry for being a noob but what am I looking at here? thanks.  Smiley

no problem at all.

yesterday, i was shorting on bitfinex during the panic sell. i saw the "bounce" off of the "ceiling" of $150 (which, from previous price data, was an obvious support/resistance). this is a bearish sign.

i opened a short, and then hunted for high-volume 'dojis'. a doji is a short candle with a small body, and a symmetrical range of high and low. when these candles pattern with a spike in volume, it is a very reliable 'reversal' indicator. this is where i closed.

i used this method twice and caught two consecutive tops and bottoms this way. is that any clearer?

--arepo

edit: the second reversal candle looks more like an inverse 'shooting star', but the small body, large volume, and large range proportional to the body are all characteristics of a reversal candle.
hero member
Activity: 826
Merit: 508
April 26, 2013, 04:18:27 PM
#43
also, here's a very simple ceiling-bounce/'doji'+volume hunting exercise i was doing yesterday while daytrading. very straightforward and effective method:

-===-



-===-

--arepo

Sorry for being a noob but what am I looking at here? thanks.  Smiley
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 03:44:53 PM
#42
also, here's a very simple ceiling-bounce/'doji'+volume hunting exercise i was doing yesterday while daytrading. very straightforward and effective method:

-===-



-===-

--arepo
sr. member
Activity: 448
Merit: 250
this statement is false
April 26, 2013, 03:39:40 PM
#41
more evidence of the rigor of some methods of TA can be found in this thread, where the OP and i came to the same conclusion based upon wildly different techniques. that's called science.

--arepo
Pages:
Jump to: