Has technical analysis been thoroughly debunked in the bitcoin community yet?

[Este Nuno]

No, I don't think that at all.

Even if you use your TA to give a price range and corresponding probabilities I still think it's no better than random unless you can statistically prove that its not.

I get where you're coming from. "If it works, it should be possible to _show_ that it works." (where "show" means something like "up to current academic publication standards"). Did I get that right?

Let me make a countering case. I suggested the analogy before in some other thread, so apologies for the repetition:

Imgagine you're a computer scientist in the 1970s. You implemented a computer chess algorithm on the fastest then-available machine, nothing more complicated than alpha-beta pruning (or variations of it). You still fail to beat a competent human GM. Consistently.

So, you ask the GM how he plays chess. He'll throw an entire library of opening theory (or endgame theory, or whatever) at you. You read it, and will quickly realize that this completely unimplementable. Worthless, from your perspective. Yet, still, that GM beats your machine nearly every single time.

End of analogy. In my opinion, TA as she is practiced by the more competent traders in here, is as much of an art as it is a science (thanks to sgbett for that phrasing). It's difficult (though not impossible) to show that it produces statistically significant results. And even then, what is tested is only a tiny subset of what traders actually use for the TA: the purely algorithmic part, while in reality the algorithmic methods go hand in hand with intuition/human-optimized pattern recognition/etc. (like in the chess analogy above).

It really runs down to the following question, in my opinion: do you only accept knowledge and results that are produced by the full rigor of academic methodology, or do you allow for "conditional knowledge"... insights that appear reasonable to you, that you have personal annecdotal evidence for, and that you hope can eventually be proven to be correct in a more formal way.

Myself, I follow the latter approach: I believe I have tentative evidence that TA (as I practice it) works well enough, but I don't expect to be able to show beyond a doubt that it does work, because (as you already pointed out) the system underlying it is too complex (and not well enough understood yet) to formalize it to the point where the question can really be answered once and for all. Until then, I will continue using those methods, with tight risk control in place to avoid catastrophic failure should the methods "stop working" for me.

Personally, I wouldn't need to accept anything close to academic rigor to even begin to accept that someone was applying a strategy that was currently working in the Bitcoin market. If someone was able to shoe me statistically that they had an algorithmic trading pattern that' produced results falling in 3 sigma/3 standard deviation range I would be more than happy to think that it's very likely that their strategy was currently producing results. I think generally in academia scientists look for 6 sigma results which is far more rigorous than I would ever need to see.

The thing that gets me the most is that if we took all the people here and on other forums who post about their TA strategies and gave them all a bunch of money to trade on a market that was actually literally just a random walk created by some programmed random function, no one would notice. They would all still use the same language, the same discussions, some people would be big winners and people would look to them for advice. It would be the same "it works till it doesn't".

"TA" can cover such a broad range of approaches that I would never dismiss everything classified as TA out of hand. But it's so vastly unlikely that anyone taking a non-quantitative layman's approach is getting information of value. Unless you know that there are enough market participants acting on the exact same "TA" and using that information to exploit their patterns. But that would never happen in any sufficiently large market.(It might have happened with the Yen in the 1980s or something like that but that's not really relevant)

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved. My whole point is that people should focus on those two issues mainly. Your edge is going to come from understanding. And the commonly used TA doesn't do anything but exploit the human predisposition to seek patterns where they are none. Patterns are a mathematical construct that can be analogized and shown to be statistically significant. Having good results doesn't indicate anything really. It's vastly more likely you're just experiencing positive variance. Not to mention the fact that anyone trading with an upward bias in Bitcoin the last couple of years was going to make money in most cases. Focusing your time and effort on other areas would probably be more productive.

[1714478123]

1714478123

[1714478123]

1714478123

[Gimmelfarb]

How about when very successful traders like Thomas Bulkowski give their statistical results? For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

http://thepatternsite.com/fallwedge.html
http://thepatternsite.com/dt.html

Is successfully trading those patterns "random"?

sing it, sister!

i don't argue often, but when i do, it's not against Mr Bulkowski. but hey, if people want to think it's all random, it's all good with me.... whatever.

[Este Nuno]

How about when very successful traders like Thomas Bulkowski give their statistical results? For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

http://thepatternsite.com/fallwedge.html
http://thepatternsite.com/dt.html

Is successfully trading those patterns "random"?

sing it, sister!

i don't argue often, but when i do, it's not against Mr Bulkowski. but hey, if people want to think it's all random, it's all good with me.... whatever.

For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

That might be the most ridiculous statement I've ever heard in regards to TA.

In what context?

And something like this would be easy to check. First define quantitatively what a "descending triangle" and a "falling wedge" is. Likewise with "upward breakout" and "downward breakout". And then back test sufficiently large set of market data.

This type of stuff in particular is the worst part of this whole TA scene. Someone making outlandish claims backed up by laughable evidence trying to sell you their expertise.

I'd imagine the average person here would find this as silly as I do. I don't think most people here are on that level of delusion. I think a lot of people here probably just put too much faith in to patterns that don't actually give any meaningful information.

[oda.krell]

No, I don't think that at all.

Even if you use your TA to give a price range and corresponding probabilities I still think it's no better than random unless you can statistically prove that its not.

I get where you're coming from. "If it works, it should be possible to _show_ that it works." (where "show" means something like "up to current academic publication standards"). Did I get that right?

Let me make a countering case. I suggested the analogy before in some other thread, so apologies for the repetition:

Imgagine you're a computer scientist in the 1970s. You implemented a computer chess algorithm on the fastest then-available machine, nothing more complicated than alpha-beta pruning (or variations of it). You still fail to beat a competent human GM. Consistently.

So, you ask the GM how he plays chess. He'll throw an entire library of opening theory (or endgame theory, or whatever) at you. You read it, and will quickly realize that this completely unimplementable. Worthless, from your perspective. Yet, still, that GM beats your machine nearly every single time.

End of analogy. In my opinion, TA as she is practiced by the more competent traders in here, is as much of an art as it is a science (thanks to sgbett for that phrasing). It's difficult (though not impossible) to show that it produces statistically significant results. And even then, what is tested is only a tiny subset of what traders actually use for the TA: the purely algorithmic part, while in reality the algorithmic methods go hand in hand with intuition/human-optimized pattern recognition/etc. (like in the chess analogy above).

It really runs down to the following question, in my opinion: do you only accept knowledge and results that are produced by the full rigor of academic methodology, or do you allow for "conditional knowledge"... insights that appear reasonable to you, that you have personal annecdotal evidence for, and that you hope can eventually be proven to be correct in a more formal way.

Myself, I follow the latter approach: I believe I have tentative evidence that TA (as I practice it) works well enough, but I don't expect to be able to show beyond a doubt that it does work, because (as you already pointed out) the system underlying it is too complex (and not well enough understood yet) to formalize it to the point where the question can really be answered once and for all. Until then, I will continue using those methods, with tight risk control in place to avoid catastrophic failure should the methods "stop working" for me.

Personally, I wouldn't need to accept anything close to academic rigor to even begin to accept that someone was applying a strategy that was currently working in the Bitcoin market. If someone was able to shoe me statistically that they had an algorithmic trading pattern that' produced results falling in 3 sigma/3 standard deviation range I would be more than happy to think that it's very likely that their strategy was currently producing results. I think generally in academia scientists look for 6 sigma results which is far more rigorous than I would ever need to see.

The thing that gets me the most is that if we took all the people here and on other forums who post about their TA strategies and gave them all a bunch of money to trade on a market that was actually literally just a random walk created by some programmed random function, no one would notice. They would all still use the same language, the same discussions, some people would be big winners and people would look to them for advice. It would be the same "it works till it doesn't".

"TA" can cover such a broad range of approaches that I would never dismiss everything classified as TA out of hand. But it's so vastly unlikely that anyone taking a non-quantitative layman's approach is getting information of value. Unless you know that there are enough market participants acting on the exact same "TA" and using that information to exploit their patterns. But that would never happen in any sufficiently large market.(It might have happened with the Yen in the 1980s or something like that but that's not really relevant)

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved. My whole point is that people should focus on those two issues mainly. Your edge is going to come from understanding. And the commonly used TA doesn't do anything but exploit the human predisposition to seek patterns where they are none. Patterns are a mathematical construct that can be analogized and shown to be statistically significant. Having good results doesn't indicate anything really. It's vastly more likely you're just experiencing positive variance. Not to mention the fact that anyone trading with an upward bias in Bitcoin the last couple of years was going to make money in most cases. Focusing your time and effort on other areas would probably be more productive.

You continue in making valid points, but you chose to not address one of my points that was intended to be a direct answer to your concerns. I selected the chess/algorithmic chess example for a reason...

First, we need to distinguish two elements here,
1) statistical evidence that some traders "beat the market" in a way that is unlikely to be the product of pure chance, and
2) determining what is the reason for 1).

I believe 1) can be shown (though your requirement that it needs to be an "algorithmic trading pattern" is not necessary for that - it doesn't matter whatever cognitive process produces the result), and there are enough well known (public) traders outside of btc that provide evidence for 1). I know some economists have tongue in cheek referred to Warren Buffet as a "six sigma event", but that's ultimately just an admission of what a sad field it is they work in: physicists would take such an event as an opportunity to investigate alternative explanations, economists (more precisely: adherents of the EMH) seem to think by labeling something that goes against their predictions an outlier they solve the problem. /rant

Next step, what causes 2), is a lot harder, and your objection is valid of course: just because a trader beats the market significantly and uses TA doesn't prove that TA is the reason for that success. Which brings me back to my chess analogy: The often somewhat fuzzy sounding strategies written over the centuries on chess might seem useless to you, coming from a purely formal position. You will still note that almost all players that are at an extremely high level will use such (non algorithmic) advice in learning and training.

What to conclude? a) their success is due some other factor (and the strategy they learn is a "placebo"), or perhaps b) there is after all some, difficult to measure, effect of those "vague strategies" that gives players (and traders) a competitive edge.

To paraphrase: I think it is entirely plausible that, if implemented purely algorithmically, most TA would produce only barely significant results. That doesn't mean however it cannot work - I find it entirely plausible that TA mainly works by "sharpening" a traders intuition for the effects you mention yourself: market psychology and underlying causes. I would add a few more "fundamental market forces" (momentum, reversion), but in principle this is where we probably agree, if phrased correctly... I don't believe TA is a set of methods that work on its own. It is perhaps a set of tools that help the trader's human mind to read markets better than is usually possible without those tools.

[Este Nuno]

No, I don't think that at all.

Even if you use your TA to give a price range and corresponding probabilities I still think it's no better than random unless you can statistically prove that its not.

I get where you're coming from. "If it works, it should be possible to _show_ that it works." (where "show" means something like "up to current academic publication standards"). Did I get that right?

Let me make a countering case. I suggested the analogy before in some other thread, so apologies for the repetition:

Imgagine you're a computer scientist in the 1970s. You implemented a computer chess algorithm on the fastest then-available machine, nothing more complicated than alpha-beta pruning (or variations of it). You still fail to beat a competent human GM. Consistently.

So, you ask the GM how he plays chess. He'll throw an entire library of opening theory (or endgame theory, or whatever) at you. You read it, and will quickly realize that this completely unimplementable. Worthless, from your perspective. Yet, still, that GM beats your machine nearly every single time.

End of analogy. In my opinion, TA as she is practiced by the more competent traders in here, is as much of an art as it is a science (thanks to sgbett for that phrasing). It's difficult (though not impossible) to show that it produces statistically significant results. And even then, what is tested is only a tiny subset of what traders actually use for the TA: the purely algorithmic part, while in reality the algorithmic methods go hand in hand with intuition/human-optimized pattern recognition/etc. (like in the chess analogy above).

It really runs down to the following question, in my opinion: do you only accept knowledge and results that are produced by the full rigor of academic methodology, or do you allow for "conditional knowledge"... insights that appear reasonable to you, that you have personal annecdotal evidence for, and that you hope can eventually be proven to be correct in a more formal way.

Myself, I follow the latter approach: I believe I have tentative evidence that TA (as I practice it) works well enough, but I don't expect to be able to show beyond a doubt that it does work, because (as you already pointed out) the system underlying it is too complex (and not well enough understood yet) to formalize it to the point where the question can really be answered once and for all. Until then, I will continue using those methods, with tight risk control in place to avoid catastrophic failure should the methods "stop working" for me.

Personally, I wouldn't need to accept anything close to academic rigor to even begin to accept that someone was applying a strategy that was currently working in the Bitcoin market. If someone was able to shoe me statistically that they had an algorithmic trading pattern that' produced results falling in 3 sigma/3 standard deviation range I would be more than happy to think that it's very likely that their strategy was currently producing results. I think generally in academia scientists look for 6 sigma results which is far more rigorous than I would ever need to see.

The thing that gets me the most is that if we took all the people here and on other forums who post about their TA strategies and gave them all a bunch of money to trade on a market that was actually literally just a random walk created by some programmed random function, no one would notice. They would all still use the same language, the same discussions, some people would be big winners and people would look to them for advice. It would be the same "it works till it doesn't".

"TA" can cover such a broad range of approaches that I would never dismiss everything classified as TA out of hand. But it's so vastly unlikely that anyone taking a non-quantitative layman's approach is getting information of value. Unless you know that there are enough market participants acting on the exact same "TA" and using that information to exploit their patterns. But that would never happen in any sufficiently large market.(It might have happened with the Yen in the 1980s or something like that but that's not really relevant)

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved. My whole point is that people should focus on those two issues mainly. Your edge is going to come from understanding. And the commonly used TA doesn't do anything but exploit the human predisposition to seek patterns where they are none. Patterns are a mathematical construct that can be analogized and shown to be statistically significant. Having good results doesn't indicate anything really. It's vastly more likely you're just experiencing positive variance. Not to mention the fact that anyone trading with an upward bias in Bitcoin the last couple of years was going to make money in most cases. Focusing your time and effort on other areas would probably be more productive.

You continue in making valid points, but you chose to not address one of my points that was intended to be a direct answer to your concerns. I selected the chess/algorithmic chess example for a reason...

First, we need to distinguish two elements here,
1) statistical evidence that some traders "beat the market" in a way that is unlikely to be the product of pure chance, and
2) determining what is the reason for 1).

I believe 1) can be shown (though your requirement that it needs to be an "algorithmic trading pattern" is not necessary for that - it doesn't matter whatever cognitive process produces the result), and there are enough well known (public) traders outside of btc that provide evidence for 1). I know some economists have tongue in cheek referred to Warren Buffet as a "six sigma event", but that's ultimately just an admission of what a sad field it is they work in: physicists would take such an event as an opportunity to investigate alternative explanations, economists (more precisely: adherents of the EMH) seem to think by labeling something that goes against their predictions an outlier they solve the problem. /rant

Next step, what causes 2), is a lot harder, and your objection is valid of course: just because a trader beats the market significantly and uses TA doesn't prove that TA is the reason for that success. Which brings me back to my chess analogy: The often somewhat fuzzy sounding strategies written over the centuries on chess might seem useless to you, coming from a purely formal position. You will still note that almost all players that are at an extremely high level will use such (non algorithmic) advice in learning and training.

What to conclude? a) their success is due some other factor (and the strategy they learn is a "placebo"), or perhaps b) there is after all some, difficult to measure, effect of those "vague strategies" that gives players (and traders) a competitive edge.

To paraphrase: I think it is entirely plausible that, if implemented purely algorithmically, most TA would produce only barely significant results. That doesn't mean however it cannot work - I find it entirely plausible that TA mainly works by "sharpening" a traders intuition for the effects you mention yourself: market psychology and underlying causes. I would add a few more "fundamental market forces" (momentum, reversion), but in principle this is where we probably agree, if phrased correctly... I don't believe TA is a set of methods that work on its own. It is perhaps a set of tools that help the trader's human mind to read markets better than is usually possible without those tools.

Okay, I think I agree with you. Using some TA can assist someone who is mindful of the fact that the TA itself is not predictive but rather tells the story up until this point in time. Considering things like support and resistance and how and why you expect the market to react to future changes is good.

But I think the vast majority of people completely misuse TA. And I think it more than likely hinders anyone who doesn't understand that it's just another tool to help view the rate of change of a price over time. At least 95% of people probably waste their time and money seeking something that doesn't exist.

[oda.krell]

But I think the vast majority of people completely misuse TA. And I think it more than likely hinders anyone who doesn't understand that it's just another tool to help view the rate of change of a price over time. At least 95% of people probably waste their time and money seeking something that doesn't exist.

Agreed.

Whenever you read something like "according to the weekly MACD, you should sell now" or "according to my log linear model, you're super safe to buy in now", feel free to ignore the advice.

(Disclosure: I have made similar statements. Oversimplification is a habit that's hard to kick entirely.)

[bassclef]

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved.

This. This is TA.

You say human action isn't quantifiable with patterns on a chart. I disagree. The field of psychology disagrees. Take a double or triple top. The price bounces of a resistance point two or three times, then falls. Now think about when you knock on someone's door. Do you knock eight or nine times? No. Two or three is sufficient to figure out there's nobody home.

To non-analysts these are simply lines on a chart. To students of TA they say a lot more, and the pictures don't actually matter, what matters is what they represent.

To use another analogy (and to argue that TA is more like an art than science), I could teach you music theory but it wouldn't necessarily make you a good musician. You would be informed, but playing Mozart or Scarlatti piano sonatas might not come easily. That takes years of practice and dedication, and even then you might have no ear at all and someone with a natural affinity to music (basically the ability to perceive it in one's head accurately without an instrument in front of them) would likely eclipse you very quickly with much less work invested. It is not the notes on the paper, it is interpreting what they represent.

In the same way TA requires good intuition, an understanding of human action, and the dedication (also obsession) of learning the markets inside and out, how/why/where they move and when, and mastering one's emotions. There are repeatable patterns that I've learned to recognize and it often feels more like art that science. I can't explain it, neither can I explain how I hear and recognize music intervals without a piano in front of me, or how I control/convey my emotions while performing. It's not very quantifiable on paper, but neither is what artists do. All I can say is that I'm able to make money doing it.

[Este Nuno]

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved.

This. This is TA.

You say human action isn't quantifiable with patterns on a chart.

No, I do say that it is.

And anyone who thinks they have the ability to do so should be able to prove it. Even if only to themselves. Tracking and honest reporting of results should be a priority for anyone.

My main point is that the vast majority of people look to TA as some sort of fancy Rosetta Stone for unlocking the secrets of markets but it's not that at all. The average person should probably ignore TA for a while until they've really gotten to the point where they understand the fundamentals of the market that they intend to trade. I've seen someone come on a forum and sell 10k TA training packs to a 10 people wanting to learn trading and seeing the subsequent embarrassment of the people who paid for it. Since then I've taken an extra hard look at what most people using TA claim and it's just such a vast amount of bullshit that the average person really has no way to tell the difference between someone just making stuff up and someone truly using a good analytic approach.

I'm not saying good traders can't make money making use of a wide range of tools. I also think that there are lots of traders who probably have a good intuitive grasp on how a market works and think that its the TA that's making them money when they would probably do just as well with out it.

[bassclef]

I'm not saying that people can't get an edge on a market by understanding the underlying causes of what moves the price, and the psychological factors involved.

This. This is TA.

You say human action isn't quantifiable with patterns on a chart.

No, I do say that it is.

And anyone who thinks they have the ability to do so should be able to prove it. Even if only to themselves. Tracking and honest reporting of results should be a priority for anyone.

I shared my results to prove something, and those results were profitable, then others would copy it making my strategy less profitable. So there is no incentive. I have nothing to prove except to myself.

I do see your other point though. There are mountains of bullshit that are marketed to new traders, promising them huge returns with no work. You don't see that too much here (yet) but in traditional markets it's a huge business that preys on the ignorant. Of course there is no free lunch.

Even sites like stockcharts.com don't go into much detail about how to apply TA. They're simply a collection of definitions... what new traders may not understand is getting the basics down is more important than understanding a bunch of indicators... it's like jumping into the deep end of the pool before you've learned to swim. Start with Dow theory--there is nothing magical about it, it's simply understanding how markets work on a basic level. Work up to support and resistance, then moving averages, then reversal signals, maybe some momentum and overbought/oversold indicators... that's pretty much all I use to time the market. Even things like Elliot Wave analysis is too much for me. Maybe someday.

[Gimmelfarb]

How about when very successful traders like Thomas Bulkowski give their statistical results? For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

http://thepatternsite.com/fallwedge.html
http://thepatternsite.com/dt.html

Is successfully trading those patterns "random"?

sing it, sister!

i don't argue often, but when i do, it's not against Mr Bulkowski. but hey, if people want to think it's all random, it's all good with me.... whatever.

For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

That might be the most ridiculous statement I've ever heard in regards to TA.

In what context?

And something like this would be easy to check. First define quantitatively what a "descending triangle" and a "falling wedge" is. Likewise with "upward breakout" and "downward breakout". And then back test sufficiently large set of market data.

This type of stuff in particular is the worst part of this whole TA scene. Someone making outlandish claims backed up by laughable evidence trying to sell you their expertise.

I'd imagine the average person here would find this as silly as I do. I don't think most people here are on that level of delusion. I think a lot of people here probably just put too much faith in to patterns that don't actually give any meaningful information.

you keep ignoring the fact that traders are using back tested data. he just pointed out an example of someone displaying their data, and went on to say his own data confirms their conclusions. that's actually about ALL one can do--stay on the side of probability until the data no longer supports it.

all evidence is laughable. there is no proof to be had. so what are you arguing here? your position is no better proven than those in favor of TA. you keep accusing people of having laughable evidence, but your own evidence that any technical analysis is no better than "random" is completely non-existent.

[ensurance982]

Or are there a lot of people still following technical indicators and such.

Of the people who still choose to follow technical analysis, what's popular in bitcoin? What are people using?

Depends on what you are using. There are pretty obscure and dubious indicators people use, but as far as I know, MACD has been a pretty stong and reliable indicator if you depend solely on it in order to buy or sell when a new trend emerges.

[Marbit]

How about when very successful traders like Thomas Bulkowski give their statistical results? For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

http://thepatternsite.com/fallwedge.html
http://thepatternsite.com/dt.html

Is successfully trading those patterns "random"?

For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

That might be the most ridiculous statement I've ever heard in regards to TA.

In what context?

And something like this would be easy to check. First define quantitatively what a "descending triangle" and a "falling wedge" is. Likewise with "upward breakout" and "downward breakout". And then back test sufficiently large set of market data.

This type of stuff in particular is the worst part of this whole TA scene. Someone making outlandish claims backed up by laughable evidence trying to sell you their expertise.

I'd imagine the average person here would find this as silly as I do. I don't think most people here are on that level of delusion. I think a lot of people here probably just put too much faith in to patterns that don't actually give any meaningful information.

Hey, thanks for deleting half of what I said and then straw-manning a single point I touched on.

I never said blindly follow Bulkowski – and it’s ridiculous for you to suggest that I was arguing that. It was a simple example of a statistical data set and how it could be used. What I said was that using back-tested systems is better than "random". I went on to say that I back-test my trading system.

You somehow turned that into a comment about snake oil salesmen. I am talking in the context of doing technical analysis -- not blindly following supposed experts.

I didn't say it proved TA works. I don't think empirical proof is possible. I said I started trading because I was successful in poker, which depends on similar statistical analysis. In my data set, Fib retracements, moving averages, certain candlestick patters, and momentum/volume flow indicators all hold statistical significance. And actually, I'm under the impression that a win rate does tell you about the EV of a given set of trades, assuming proper risk management.

Comparing TA that incorporates significant backtesting to "calling a naked gutshot on the turn" is empty rhetoric. That's plain for all to see.

I'm not saying good traders can't make money making use of a wide range of tools. I also think that there are lots of traders who probably have a good intuitive grasp on how a market works and think that its the TA that's making them money when they would probably do just as well with out it.

http://en.wikipedia.org/wiki/Argument_from_ignorance

Even if you use your TA to give a price range and corresponding probabilities I still think it's no better than random unless you can statistically prove that its not.

http://en.wikipedia.org/wiki/Philosophic_burden_of_proof

That's just like, your opinion, man.

[oda.krell]

Or are there a lot of people still following technical indicators and such.

Of the people who still choose to follow technical analysis, what's popular in bitcoin? What are people using?

Depends on what you are using. There are pretty obscure and dubious indicators people use, but as far as I know, MACD has been a pretty stong and reliable indicator if you depend solely on it in order to buy or sell when a new trend emerges.

MACD, daily, is probably the most "vanilla" indicator used among btc traders, at least that's my impression. And it actually performs reasonably well, most of the time... let_me_backtest_that has thread in here where he backtested a "pure" MACD strategy, and the results weren't half bad, _especially_ if one considers that it reduced the max drawdown of your USD account substantially. In other words, even if it doesn't greatly outperform buy and holdin profits, it could be used to "protect" the USD value of your trading account. (Note: I don't really recommend anyone uses a pure MACD strategy).

[ensurance982]

Or are there a lot of people still following technical indicators and such.

Of the people who still choose to follow technical analysis, what's popular in bitcoin? What are people using?

Depends on what you are using. There are pretty obscure and dubious indicators people use, but as far as I know, MACD has been a pretty stong and reliable indicator if you depend solely on it in order to buy or sell when a new trend emerges.

MACD, daily, is probably the most "vanilla" indicator used among btc traders, at least that's my impression. And it actually performs reasonably well, most of the time... let_me_backtest_that has thread in here where he backtested a "pure" MACD strategy, and the results weren't half bad, _especially_ if one considers that it reduced the max drawdown of your USD account substantially. In other words, even if it doesn't greatly outperform buy and holdin profits, it could be used to "protect" the USD value of your trading account. (Note: I don't really recommend anyone uses a pure MACD strategy).

Yeah, I'm talking about that backtesting-example/experiment! It actually is a vanilla indicator around here, I think that expression is exactly spot on when we're talking about MACD But much like Occam's Razor always suggests: It isn't always wise to use overcomplicated things.

[Este Nuno]

How about when very successful traders like Thomas Bulkowski give their statistical results? For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

http://thepatternsite.com/fallwedge.html
http://thepatternsite.com/dt.html

Is successfully trading those patterns "random"?

For example, a falling wedge results in an upward breakout 68% of the time. A descending triangle results in a downward breakout 64% of the time.

That might be the most ridiculous statement I've ever heard in regards to TA.

In what context?

And something like this would be easy to check. First define quantitatively what a "descending triangle" and a "falling wedge" is. Likewise with "upward breakout" and "downward breakout". And then back test sufficiently large set of market data.

This type of stuff in particular is the worst part of this whole TA scene. Someone making outlandish claims backed up by laughable evidence trying to sell you their expertise.

I'd imagine the average person here would find this as silly as I do. I don't think most people here are on that level of delusion. I think a lot of people here probably just put too much faith in to patterns that don't actually give any meaningful information.

Hey, thanks for deleting half of what I said and then straw-manning a single point I touched on.

I never said blindly follow Bulkowski – and it’s ridiculous for you to suggest that I was arguing that. It was a simple example of a statistical data set and how it could be used. What I said was that using back-tested systems is better than "random". I went on to say that I back-test my trading system.

You somehow turned that into a comment about snake oil salesmen. I am talking in the context of doing technical analysis -- not blindly following supposed experts.

I didn't say it proved TA works. I don't think empirical proof is possible. I said I started trading because I was successful in poker, which depends on similar statistical analysis. In my data set, Fib retracements, moving averages, certain candlestick patters, and momentum/volume flow indicators all hold statistical significance. And actually, I'm under the impression that a win rate does tell you about the EV of a given set of trades, assuming proper risk management.

Comparing TA that incorporates significant backtesting to "calling a naked gutshot on the turn" is empty rhetoric. That's plain for all to see.

Yes, I did forget to reply to the other part of your comment. Sorry. I got so caught up in addressing that website that it slipped my mind.

I wasn't using the poker analogy to compare to someone who uses proper statistically significant back testing  I've been trying to say that anyone who is not doing such things is very unlikely to be doing any better than random.

I didn't see any evidence on the guys site other than the fact that he claims all data is based on "perfect trades". Which makes me extremely suspicious that he's data snooping.

I'm not saying good traders can't make money making use of a wide range of tools. I also think that there are lots of traders who probably have a good intuitive grasp on how a market works and think that its the TA that's making them money when they would probably do just as well with out it.

http://en.wikipedia.org/wiki/Argument_from_ignorance

I'm not making any assertion here. It's not something I'm trying to prove, I just suspect that many people aren't benefiting as much as they think from using anything other than the very basics of TA.

Even if you use your TA to give a price range and corresponding probabilities I still think it's no better than random unless you can statistically prove that its not.

http://en.wikipedia.org/wiki/Philosophic_burden_of_proof

That's just like, your opinion, man.

Well I assume you would agree with me that anyone claiming extraordinary results should be able to provide statistically significant proof, no?

But to start with anyone describing their predictions in terms of a probability distribution is already doing much better than the average person misapplying TA.

[Marbit]

That's fair enough. I can appreciate that.

[BowieMan]

Or are there a lot of people still following technical indicators and such.

Of the people who still choose to follow technical analysis, what's popular in bitcoin? What are people using?

I think the back-testing of various TA indicators is very interesting! If you take some well known indicators, you may really see that they can be helpful in doing informed decisions. But Even though TA is somewhat of a self-fulfilling prophecy, I think a lot of people are putting way too much faith in it!

[wasserman99]

TA is somewhat of a self-fulfilling prophecy

This is something I hear a lot. I'm curious -- what makes people so sure of this assumption? It's not exactly something that one could provide any evidence for.

[BowieMan]

TA is somewhat of a self-fulfilling prophecy

This is something I hear a lot. I'm curious -- what makes people so sure of this assumption? It's not exactly something that one could provide any evidence for.

Yeah, it's difficult to prove, but: If a lot of people use common TA indicators (for example MACD, which is a very common one), then they trust that indicator. If the MACD now turns green or goes above a certain value, they buy., and others will as well - since they are using the same indicator (common ones). On the other hand, people also expect other people to act according to those indicators, and therefore anticipate their buys, and try to buy even earlier. It's somewhat of a self-fulfilling prophecy.