I have attended a few TA workshops, a few instructors said: 1) If you are the only one see the signal/price pattern/indicator, your TA is no going to make any money The herd is not going with you. 2) If everybody see the signal/price pattern/indicator, you will not be making any money. You are in the middle of the herd. You have absolutely no edge. 3) If you are in the group of traders who see the signal/price pattern/indicator together and trade relatively early, you may make money. You must be the few leading sheep ahead of the herd and not the last of the herd going in the same direction. 4) If you change the time frame, the signal/price pattern/indicator would mean entirely different. 5) Different traders can have different conclusion and take different action when looking at the same signal/price pattern/indicator. If TA work better than 68% or 2/3 or the time, it is not any better than the one normal standard deviation of price movement. TA is all up to interpretation of the individual trader. TA cannot be proven mathematically to be better than 50%. It TA win better than 50%, you would have an edge, all the big money investment firms would hire programmers to do it with computers. You got 50% with coin toss. The TA information is open to everyone. There is no secret, no edge.
"statistically significant" when used in this context infers that the odds are greater of one outcome over another. correct, it certainly does not mean a sure thing. i prefer to use statistical vaild concepts when trading, over those that are not. regards, surfer
Pardon? This is bad ... why (ignoring that its a mixed up statement invoking normal statistics with joyous ignorance)? If any new trader can win 68% of the time (hell or even 66.666* percent of the time) then unless his losses are bigger than his wins he can become a happy little trader. Very happy.
not everything is scalable. if you trade at 100% win rate, but only trade once a day making 100$ a trade. you are below the poverty line.
No. You misunstood my comment. I did not agree that TA can be as good as 68%. Using one standard deviation (68%) base on option implied volatility , one can approximate the price range, not direction within a certain period. TA cannot predict price movement direction or range. Nothing can predict price movement direction. It is either up, down, or none. And, of course when? Where is your statistics you can offer to prove TA along, without trader interpretation can be reliable better than 50% .
Before you condemn all of those indicators, find out if Aronson used a logical position sizing method in his tests. Almost no TA books even mention position sizing and even a few so-called trading books commit that mortal sin of omission. A viable trading strategy isn't just rules on when to trade, it must also dictate how much to trade. I'd be very interested in learning if Aronson's tests used a plausible method of position sizing, because even the best timing methods can be defeated by overbetting.
That is a great quote. Honestly reading the traditional literature isn't really that hard. If TA were successful then you're right, investment firms would hire programmers to do it with computers. Or are these people who are beating the markets using TA going one step further where they have an edge. You also make a great point, if information is open to everyone, there is no edge. Now just obtaining that information that is still edgeable...well that's the hard part.