TA does/doesn't work thread with a difference.

[enragedcow]

found this while playing online game

this item's price chart has quite a clear head and shoulders pattern and it is fair to assume that very few of the market participants (13 - 30 year olds) would have any knowledge of technical analysis



so does TA work? probably

 

[Stirf]

found this while playing online game

this item's price chart has quite a clear head and shoulders pattern and it is fair to assume that very few of the market participants (13 - 30 year olds) would have any knowledge of technical analysis



so does TA work? probably

This is the H&S pattern from July 2009 a very clear one in the S&P as well (see thumbnail). We all know what happened after July 2009. The market kept on rallying untill it's high from last April. So does TA work? Probably not.

Took this chart from the net so I apologize for the crappy technical indicators in the attachment. Besides this, what if the Nasdaq, HangSeng, Nikkei or whatever market didn't show a H&S pattern? Would they still go down because the S&P is the leading market? Who/what decides what market is leading? do all the indices plunge because of a H&S pattern in one of them? What if you change your settings from e.g. hourly bars to daily bars or even weekly bars? Will there still be a clear H&S pattern?

TA is one big joke with a great marketing unit behind it :cheesy:!!

 

[scose-no-doubt]

Well was kind of the point of the thread when I started it. there are a million thread on how great TA has made people hella money and why but none that show where people lose money and why. how can you asses something by looking only at one end of the spectrum of results? Stupid.

This is the H&S pattern from July 2009 a very clear one in the S&P as well (see thumbnail). We all know what happened after July 2009. The market kept on rallying untill it's high from last April. So does TA work? Probably not.

The H&S was there and was failed and was rejected BUT if we are to believe that it was rejected then do we not have to accept that it recognised by the market? Does that in itself not constitute that TA works or was it just that JP's earnings gave some confidence to the market and stocks were cheap relative to expected dividend yields? Was it a combination of both?

Took this chart from the net so I apologize for the crappy technical indicators in the attachment. Besides this, what if the Nasdaq, HangSeng, Nikkei or whatever market didn't show a H&S pattern? Would they still go down because the S&P is the leading market? Who/what decides what market is leading? do all the indices plunge because of a H&S pattern in one of them?

Well I always assumed that this is a factor of what markets are open, global/regional growth prospects relative to price/yield and the non arb principle. Some stocks are even listed multi exchange innit

What if you change your settings from e.g. hourly bars to daily bars or even weekly bars? Will there still be a clear H&S pattern?

I don't see why this bit is relevant. if teh patternz occur in a timeframe i can has it assessed in same timeframe

Side discussion . Are non-daily timeframes a load of tosh?

 

[Pazienza]

Well was kind of the point of the thread when I started it. there are a million thread on how great TA has made people hella money and why but none that show where people lose money and why. how can you asses something by looking only at one end of the spectrum of results? Stupid.

Side discussion . Are non-daily timeframes a load of tosh?

It's a good point. The simple answer to the question "Does X work?" is "No", because nothing works. Realising that is a big step towards profitability.

Side discussion: Lower time frames.

In my view they are essentially no different, I consider it more of a question of a difference in scale. You will see the effects of people shunting the market around more on lower time frames, although this is not necessarily to your disadvantage. You either need to recognise what is happening and act accordingly (fairly tricky in my opinion) or adapt your trading such that the effect of this is likely to be minimised.

I think the trouble with the lower time frames is really that they offer too many "opportunities" each day, and this tends to increase the chance of over-trading, loss chasing and so on, and this is one of the main reasons for failure.

I think the thing with intraday is that everyone wants to do it, and it is very hard indeed when you are starting out. But basically just requires experience and "backbone" - the old-fashioned virtues of patience and discipline.

 

[Stirf]

Well was kind of the point of the thread when I started it. there are a million thread on how great TA has made people hella money and why but none that show where people lose money and why. how can you asses something by looking only at one end of the spectrum of results? Stupid.




Side discussion . Are non-daily timeframes a load of tosh?

"The H&S was there and was failed" FULL STOP
"BUT if we are to believe" as far as I'm concerned you can believe anything, I don't care. I stick to the facts and that is that "the pattern" failed.
"Some stocks are even listed multi exchange innit" True, but although economies are linked to eachother in many different ways that doesn't mean that when Greece has a bad debt record, Germany has it as well. Or does it ?
I don't see why this bit is relevant. if teh patternz occur in a timeframe i can has it assessed in same timeframe. You have a look at your MACD in a 1 min candlestick chart and have a look at the same MACD in a 60 min.candlestick chart. No I'm not even talking about using a point and figure chart or a 12 tick break out chart in the comparising chart.

 

[scose-no-doubt]

"The H&S was there and was failed" FULL STOP
"BUT if we are to believe" as far as I'm concerned you can believe anything, I don't care. I stick to the facts and that is that "the pattern" failed.
So if you trade a H&S failure are you not trading TA?
"Some stocks are even listed multi exchange innit" True, but although economies are linked to eachother in many different ways that doesn't mean that when Greece has a bad debt record, Germany has it as well. Or does it ?
But stock and bond markets are different animals. I don't think that's a particularly good comparison.
I don't see why this bit is relevant. if teh patternz occur in a timeframe i can has it assessed in same timeframe. You have a look at your MACD in a 1 min candlestick chart and have a look at the same MACD in a 60 min.candlestick chart. No I'm not even talking about using a point and figure chart or a 12 tick break out chart in the comparising chart.Yes but can that not be summed up in that (i believe) that market prices tend to be more random in the short term due to the order driven nature whereas the longer timeframes will be more in tune with the prevailing bias?

 

[Stirf]

So if you trade a H&S failure are you not trading TA? As far as I'm concerned you can trade on the angle of the moon vs earth or a bunch of tarot cards. But that doesn't mean it works.
But stock and bond markets are different animals. I don't think that's a particularly good comparison. You're allowed to think that, but you get the general idea. Otherwise you wouldn't talk about comparison.
Yes but can that not be summed up in that (i believe) that market prices tend to be more random in the short term due to the order driven nature whereas the longer timeframes will be more in tune with the prevailing bias? Not realy. There are traders that work with a short time frame (day traders) and there are traders that work with a longer time frame, big institutions. Both work just as random. The difference is however that the big institutions seem to be working less random because their trades usually take longer. Their profits are larger, but so are their losses. While as a daytrader might take 3 to 10 trades a day. That makes the broker very happy! But with smaller losses and smaller profits. But both groups in this case look at completely different charts. And within those two groups there are traders that work with point & figure charts, candlestick charts, x tick break out charts, using different indicators with different settings, etc. Everybody sees something different.

 

[Stirf]

Here's chart of the Dow Jones over 105 years, so quite a long time frame.

Hell if you look at the far right of the chart you might even be able to spot a reverse head and shoulders pattern! So what's it gonna be? No one knows :cheesy:!

 

[neil]

What a pointless discussion
You can use rune stones if you like or chicken bones; The price is either ranging; going up or down. Your roll of the dice or whatever will not move the market. TA is merely a mental crutch to give you the courage to make a decision. Get it wrong and hopefully you get out. Keep it simple

 

[scose-no-doubt]

Your roll of the dice or whatever will not move the market

Misses the point a bit.

TA is merely a mental crutch to give you the courage to make a decision.

So you think TA is nonsense then?

 

[jimioddy]

I’ve been getting emails about this thread recently so dared a comment:
The very name ‘Technical Analysis’ suggests to me that TA is not a subject for belief, but for proof.
It’s pretty straightforward to arrive at a quantitative measure of whether an analytical technique ‘works’ or not using a computer program. Plenty of people do it, as do I. Did I want to learn how to program? No way! But I now look at a chart from a very different viewpoint.
The answer to “Does TA work” is not yes or no, nor maybe – but … “for this period of time, for this currency pair (or whatever), this technique, using these parameters gave this result”.
Optimisation software gives you the ‘exact’ parameters (to many decimal places), which delivered the best, often outstanding, result over a past period. But of course this is a kind of ‘curve fitting’ with no guarantees for the future.
My approach is to respond to what is happening, not to try to predict it: Write a technique, test it over a past period, and see how it works in real-time using pretend money. Then go for it if the results are good enough for long enough.
Has there been any threads on this approach?

 

[scose-no-doubt]

Has there been any threads on this approach?

Maybe trader dante's pin bar threads a few years ago but he robbed it off of James 16 who subsequently (allegedly) turned out to be a shyster.

I think this is what people tend to do in their own time when they develop their own strategies but I am yet to see anyone record the development process in a blog.

Tell you what if you end up something workable I reckon you'd easy have more than a handful of of £50pcm punters for your efforts.

 

[DionysusToast]

just gonna be a troll and point your attention to the strong form of Eugene Fama's EMH that all information is priced into asset prices rapidly thus no individual can outperform the market

then i'm going to make my economics and accounting levturers angry by saying "******** mr Fama!"

'tis my belief that technical analysis of price charts does in fact lead to above market returns (gets ready to be chased off campus by enraged lecturers)

Yeah - EMH is quite amusing if you think about it.

Look at a news release & how price moves. If we listen to Fama, we think it's because there's a bunch of clever people in rooms at all the funds who put the news into their formula's, come out with a price which they all agree on and that is where price ends up. Then this process is repeated all day by these people pricing information in. This is a far cry from the reality of getting in, seeing price move against you, panicking, getting out and then seeing price move back the way you expected it to. Or is that just me?

One of the failings of TA is also one of the reasons people say TA works so well. A bunch of people entering the market all the time AKA a "self fulfilling prophecy".

Let's say you had a couple of billion dollars and could nudge the market around. If you saw a place where many people went short, why would you not put your money against there's to benefit when they puke out? By definition - if many people go short in a certain place, then their exits will be just above. This 'pool of liquidity' created above their entry is a magnet for price.

Take that a step further - let's say you are a fund and you want to go long. You have 2 choices - just start buying with market orders OR entice other people to sell to you. You can see the attraction in the latter - better prices, less visibility. Of course, you could choose to believe the markets are too pure or too large for this sort of manipulation to go on.

If you believe that the manipulation does go on, then where, as a manipulator would YOU choose to fake people out? Support/Resistance points? Trendlines? 50/200 daily MA?

 

[timsk]

I always struggle with these discussions about whether TA 'works' or not. The inference is that some traders treat it like a machine such as a TV. If you turn on the TV but fail to get either sound or picture then, all other things being equal, you conclude that it doesn't work. If this is your approach to TA, then yes, I can see that most of the time it won't work for you. It's all about what you want and expect TA to do for you and how it will assist your trading.

TA can't be all things to all traders. Take a double top. One set of traders will be looking to play the break out to new highs, while another set are looking to play the reversal. One or other of them - and possibly both - are going to be disappointed. For the losing traders to then say 'TA failed - it doesn't work' - is akin to blaming the messenger for an unwelcome message. A chart is merely a record of price movement and TA is the analysis (which is often very poor) of that chart. How you use that information and the expectation you place on it is down to each individual trader. And there's ample evidence littered all over these forums that for some traders TA 'works' extremely well. Look no further than bbmac on this very thread. Think on this: let's say that you identify a pattern that only 'works' 1% of the time. But, you're able to identify the other 99% that it doesn't work so you don't take those trades and only take the one winning one. Suddenly the 1% pattern looks rather attractive don't you think!

Tim.

 

[the hare]

I always struggle with these discussions about whether TA 'works' or not.

The problem is most people discussing technical analysis on public forums believe its predictive. The tiny minority who actually get around to robustly testing the predictive capabilities reach the inevitable conclusion that TA does not work. The real conclusion is that TA has no predictive capabilities (as if trading somehow required such nonsense)

This is a great example of the damage forums such as T2W inflict on new traders.

 

[barjon]

mmm, I doubt that many people who read forums like this carefully would conclude that TA is predictive (unless all they look at is the "vendor" posts before we get around to deleting them ), but many would conclude that TA gives them a useful basis for making their trading decisions.

Trading is an extremely simple process. You enter the market and then you exit the market. Both of those simple steps require a decision on your part.

It seems to me that most discussion on TA concentrates on the entry decision and if that floats your boat then so what - you've got to make the decision somehow. For me, I've a fairly logical mind so I like to think there is an "order" in things and TA gives me a sort of comfort blanket that my entry is "rational". I do, however, have the wit to know that nothing is certain and that my "rational" entry may well go belly up however well similarly timed entries may have fared in the past. To me, getting in is the easy bit.

It's where you get out that determines whether you put bread on the table or not. I don't think I'm alone in thinking that this is by far the most difficult decision. After well over thirty years I still fight those twin evils of reluctance to take a loss and over-keenness to snatch a profit. In theory, TA, should give me a similar comfort blanket for my exits decision, but mostly it goes out of the window.

good trading

jon

 

[0007]

The problem is most people discussing technical analysis on public forums believe its predictive. The tiny minority who actually get around to robustly testing the predictive capabilities reach the inevitable conclusion that TA does not work. The real conclusion is that TA has no predictive capabilities (as if trading somehow required such nonsense)

This is a great example of the damage forums such as T2W inflict on new traders.

As a generality I completely agree with you. However, the cycle analysis work of Hurst and Millard shows that in certain identifiable cases (probably about 10% - and I'm talking mainstream stocks here, not the type of instrument that bucks & turns at the slightest provocation) it is possible to be predictive with a high degree of reliability.

But this won't help many people because it's not straightforward and you still require all the discipline and perseverance essential for any form of successful trading.

 

[0007]

........... let's say that you identify a pattern that only 'works' 1% of the time. But, you're able to identify the other 99% that it doesn't work so you don't take those trades and only take the one winning one. Suddenly the 1% pattern looks rather attractive don't you think!

Tim.

Good point. And it works.

 

[neil]

Misses the point a bit.

So you think TA is nonsense then?

Originally Posted by neil
Your roll of the dice or whatever will not move the market

Misses the point a bit.

Substitute " moving averages etc"
Quote:
Originally Posted by neil
TA is merely a mental crutch to give you the courage to make a decision.

So you think TA is nonsense then?
I didn't say that - pay attention

 

[Stirf]

Think on this: let's say that you identify a pattern that only 'works' 1% of the time. But, you're able to identify the other 99% that it doesn't work so you don't take those trades and only take the one winning one. Suddenly the 1% pattern looks rather attractive don't you think!

Tim.

Are you trying to say you don't have losing trades?