Visit our COVID-19 site for latest information regarding how we can support you. For up to date information about the pandemic visit www.sacoronavirus.co.za.

bs-regular
bs-extra-light
bs-light
bs-light
bs-cond-light-webfont
bs-medium
bs-bold
bs-black

Community


Share knowledge. Ask questions. Find answers.

Online Share Trading

Engage and learn about markets and trading online

How accurate is TA

Reply
Salv
Frequent Contributor
I dont know much about TA, still waiting to get a booking that the next TA lecture. My question though is for example if you read the Standard Bank TA under buy sell ideas they makecertain "predication" both up and done, therefore based on TA how accurate can ones "predictions" be ?
0 Kudos
18 REPLIES 18
Not applicable
I am still a novice at TA, but there is definately good value in doing it. Since I've been on the course, could short or long the market with much more confidence, and so far starting to pay off. I believe one, need to look at both fundamental's as well as TA, combined. But TA is still a subjective manner in which one look at the market, so if you learn about peoples emotions, the TA tool work well. OMHO
0 Kudos
richardw
Super Contributor
It's not likely I'll make many friends with this comment, but I think it's akin to reading tea leaves. I'll look at it, and it'll give me an idea of what some people are thinking (which can be useful), but don't feel bad if you find it a bit suspicious that 2002's stock price will somehow be anything but one factor amongst a hundred others. Absent any other drivers, these things might come to pass. Check your astrological charts for another input :> (The moon is coming under Jupiter's influence. Sell SOL.)
0 Kudos
Ninja
Super Contributor
Trading without TA is like trading in the dark or for a long term horizon. Overbought and oversold indicators confirm pivot points, momentum indicators confirm trend strength and this combined with intra market analysis helps to make a fairly secure trade. I would never trade futures or for the short term with TA.
0 Kudos
Ninja
Super Contributor
sorry....I meant I would never trade futures or for the short term without TA.
0 Kudos
Not applicable
Many academics and similar people dismiss TA out of hand. The best mathematical model for the market is Brownian motion, which means that at each moment, the direction and magnitude of the movement is random. This model would be entirely accurate if all of the market participants were random number generators. However, being human, they do not always act randomly, hence the Brownian motion model is not perfect. TA is an attempt to quantify the psychology of the market - the participants' non-random decisions - and trade accordingly. It does obviously have limitations, e.g. different underlying psychology could lead to similar patterns; not all investors think the same, etc. These limitations make the predictions somewhat probabilistic, but still better than random. Note that if the market really is Brownian motion, then no-one would be able to make money from short-term trading - you would have equal chances of winning or losing, and the commisions would slowly eat into your capital. That there are traders that make money consistently points to the fact that TA does have value.
0 Kudos
richardw
Super Contributor
I don't think that it's a binary issue (random walk or not). That the market isn't fully random does not mean that TA Must Work - proof often requires a more rigorous link. If everyone is using the same TA programs I can see how they'll lead each other around by the nose, absent any other info. In any case - it quickly becomes a religious discussion and my only aim is only to confirm to the original poster that there are alternative views.
0 Kudos
Not applicable
Richardw, the fact that there are traders who make money consistently using TA alone is a proof that it can work. Any other way of assessing the psychology (non-random tendencies) of the market could also work - but please list these alternative methods. Your original suggestions of reading tea leaves and astrology are probably not viable - I have not heard of any traders that make money consistently using these methods. The reasons why TA works are not important - you can make money from it regardless. I also believe that there is an aspect of self-fulfillment in it. However, there are a number of ways in which human perceptions of randomness and risk are distorted; and conventional wisdom often fails to be logical, particularly when emotion is involved. Such aspects surely also play a part.
0 Kudos
SimonPB
Valued Contributor
van tharp made money using a dart and a coin. it is about managing risk and selling well. The buying is largely unimportant.
0 Kudos
Not applicable
Forgive my ignorance, but how did he know when to sell?
0 Kudos
SimonPB
Valued Contributor
trailing stop of 2% on stocks. Wait for stop to be hit.
0 Kudos
Not applicable
Thisd could imply the existence of trends, in which past upward price movement implies a greater than even probability of continuation upward. This is not the case with pure Brownian motion. This would then confirm that those portions of TA that deal with trends could make one money. It might be, though, that he was just lucky to do his experiment in a bull market? A strong demonstration of the existence of trends would require profits in both a bull and bear market, or an equal number of long and short trades in any market.
0 Kudos
richardw
Super Contributor
I think you're right that TA is linked to the psychology of the market, in relation to where it's been. So that's why I say *assuming no other information*, it could work. You're trying to out-guess the other very short term participants, who IMHO are largely using the same tools as you. So yeah a head-and-shoulders will produce an effect specifically because everybody jumps on it. But if GM goes under, I don't need a TA program to tell me the direction of the market on Monday :) Any prediction any program can make can be trounced by info coming out five minutes later - thus "absent any other information", which makes a given tool a bit too unreliable for my money. On TA working, I'd say the money made by good TA'ers is provided by new TA'ers. Come on in, the water's fine! Pay no mind to the fins swimming around you.
0 Kudos
moomoo
Contributor
That depends on what you buy and when you buy it. Obviously you cant go buy high in a market that is coming down and expect great rewards unless you an extremely patient investor. However if you are a trader, you need TA and you need some sort of trend to guide you on the path to follow. Without that trend, it's like burning money in the dark.
0 Kudos
richardw
Super Contributor
Really random behaviour often produces runs in a particular direction - it has to, or it wouldn't be random. Given any bet, some of those will produce wins, some losses. So how do we improve our odds? If you cut the losses short and take advantage of the wins, you can still do okay if your wins outweigh your smaller losses + trading costs. If you do that on extreme situations where reversion to the mean can give you better-than-even odds, I think you improve your odds. If you're sure of a medium-term trend so you know what direction to bet on, I think you improve your odds. Sans relying primarily on market psychology.
0 Kudos
Not applicable
Most of the market participants are longer-term. They might be the kind of folks who would wake up in a panic one morning and think "Oh no, XYZ is going down the tube, I'd better sell my shares," and thereby cause a spike down on the open with a good probability of the price recovering once the panicked sellers have sold. This might be one possible underlying reason why a gap often closes. There is a great leap of abstraction between the underlying psychology and the charts, so perhaps my assumptions on the underlying psychology are wrong. But I think that the good traders take money from both the bad traders and the longer-term participants. Remember most of the shares are held by funds; their holdings are too large to sell in an instant, so they are sitting ducks for traders who can act quickly. Note also that most funds underperform the index; the missing profits go to the traders. I must say, I don't pretend to understand the psychology behind the head and shoulders formation at all; I've got a very tenuous theory that people like symmetry and so they assume that the price is likely to follow the path that would create a nice symmetrical chart (this would also apply to the double top, etc.). Take note that this would not just be TA as a self-fulfilling prophecy; it includes the TA non-believers who look at the chart (why would they look at the chart if they don't believe in TA - it is illogical, but many do) and subconciously assume that the symmetrical outcome is more likely. Most likely my theory on this is completely wrong, though, so please don't tell anyone you heard it from me.
0 Kudos
Not applicable
Assuming Brownian motion, at any point, the price can move up or down with equal probability, irrespective of previous movement. Sure, trends will sometimes become apparent, but even after such a trend, the probabilities of moving up or down will be equal. It's like tossing a coin many times - the outcome of the next coin toss does not depend on the outcome of the previous tosses - it is always 50/50. Because future movement does not depend on past movement, any time is as good as any other to buy. So imagine that you always enter your next trade exactly as your previous trade was stopped out (this point in time is as good as any other). Now, effectively, your multiple small trades have become one large one, with equal probability of being up or down, but you paid commission for each individual trade. You can't consistently outbet Brownian motion.
0 Kudos
richardw
Super Contributor
Yes, but the market isn't brownian. If you flip a coin, each flip has no relationship to those that follow it. But stock movements are related because companies have a value. Move too far from that value and it'll revert at some point because someone will say "uh...guys...Pets.com isn't worth that much". Shorter timeframes produce more random motion - the market reacts to the tiniest (sometimes fairly irrelevant) bit of information. Enough of that and you get irrational movements, which you can take advantage of assuming you choose a situation where you can stay solvent longer than the market stays irrational!
0 Kudos
Not applicable
OK, I will have a crack at the head-and-shoulders psychology. A share price is rallying and it retraces - this is normal. It then rallies to new highs and retraces again. Nothing wrong here, but now there is real interest since it is setting higher highs. It rallies a 3rd time, but doesn't test new highs - clearly nothings going to happen and the bulls get out and look for something else, now you have oversupply (bulls trying to sell) and no demand, and a short term selloff.
0 Kudos