Quote from hans37: View of a newby. To me TA must be an art form because if it was objective enough for many traders to use identically the profitability would evaporate. LOL, I'll bet you are not so cavalier with broacasting your trading edge. Hint Hint get my drift? The only one posting stupid crap is you.
Jack, my I suggest rather than "stupid facts" you replace that with "misperceptions". That would be more palatable and reduce verbally induced defense mechanisms that cause barriers to learning.... Best Regards Oddi
The biggest basic flaw in TA is........ INCONSISTENCY!!!!!! There is not a single indicator in public existence that works with anything resembling consistency. This fact is only ascertained in real time, on the hard right edge of the chart. Delayed data will always show that this MACD crossover worked, or that Stochastic divergence works. Funny how this is usually not the case in real time, which is why most guru's cannot make real time calls. Even if that guru makes his call 30 seconds later, it is still DELAYED data!! Furthermore, often the guru enters his position, makes his call and hopes that enough acolytes push HIS position into the black on a very short term horizon. Three ticks is more than enough to make this strategy profitable. So guru gets out, the acolytes get gunned.
http://www.nber.org/papers/w7613 Read it, and weep. For those who don't want to read it, let me summarize. Some geniuses from MIT did a study on technical analysis within Nasdaq. The results were startling. The most bullish signal, the inverse head-and-shoulders pattern, produced an average 4% increase in the price of a stock on the third day after the pattern's completion. The most bearish signal, broadening bottoms, produced an average 6.2% decrease (charts). The authors also looked at other statistical measures such as standard deviation and skew and found that they also were significantly different from those of a randomly chosen day in the market. Let me add that I don't believe Technical Analysis is a crystal ball. In fact lots of times it's wrong. It's the way technical analysis allows you to control your emotions in respect to the stop loss that makes technical analysis so very sucessful imo.
I agree with the person who commented that my use of stupid facts is not a good descriptor. It was used to start the thread and I just wanted to go with the flow at my initiative. Not too good. So for the above posted comment, my view is that MIT did what they did and got the result they got. this is an example of using a data base (NASDAQ) that is ill chosen to do anything in particular. It suggests that the MIT team was looking at a few formations that were known to produce results and they went into a study to find out something that appears to be price changes that have some relation to the formations (completed ones, maybe) and what happens afterwards. This all fits under the heading of a book entitled "How to Lie with Statistics." Were they to sharpen their pencils and get down to repeating the approach of W. J O'Neil who endeavored to "discover" "patterns"and go further to determine their distribution in the spectrum of possibilities using percentile groupings of things like EPS and RS where these are percentiles of earnings and price performance respectively. So the "cup and handle" came into being from this. Well here we are in misconception land again regarding the MIT post. Where do the patterns and formations appear in the MIT study. For sure we know that WJO was working the EOD turf and the C&H is an intermediate term pattern or formation. It is usful on EOD data and the C&H takes about 13 weeks to form. The BO out of the handle takes an intermediate term trend to fulfill its result. Mr Market used to post here on stuff. One of the alternativesin the same sector to one of his unsuccessful picks was a stock that was recommended to him for consideration. That IT run up on that stock was from 10 to 80 over the run in about a year. TA performance is very very significant when it is done in a purposeful context. When it is studied as MIT did is is just something done on the misconception level. MIT's misconception is that you can apply TA to the NASDAQ. You can't. TA really makes money for people if they chose an excellent quality universe for applying TA. MIT missed the boat. W. J. O'Neil on the otherhand did an excellent job of creating a pattern or formation and then went about defining the universe to which it best applied. It turns out all patterns that have names follow the P, V relationship, which in itself is not completely expressed mathematically. The addition correlary fixes that. when one has a set of patterns that all follow a rogorous statement, then the person is in a place where he can begin to find an application that makes a lot of money. Having the specific narrow application arena down cold, then the person can begin to address when to make money with the application. This is the issue of "rotating" capital through the sequentially appearing investment opportunities that cycle along for the trader. If the three primary market variables cycle with periodicities that are low integer multiples, how can it be that a TA practitioner cannot understand what is going on? From this there is no viable argument that prediction is a requirement of making money. Why would anyone work on algorithms for making money? Its like someone had the misconception of looking at potential annual ROI's and concluded that one person would soon take in all the money out there. For traders, there is no connection between one and the other. Its like the miconception that markets are not changing in size. What was the fastest growing financial factor that Warren Buffet dealt with in his career? It sure wasn't his ROI.
Very interesting and logical way of looking at it. Never saw it that way, really. Nice post... Cheers.
With regard to using TA in a non lagging manner(and this includes indicators) it is best to show the future on your display. No one should have the forming bar on the very right side of their display. There is a wonderful use of the space on the right side of the display. It is the best place to do annotations of price, volume, indicators and general verbal comments. 1/3 of a chart is a good portion. If you use horizontal rays on volume, you get a nice picture of the possibility of price movement. Three rays are a possibility. One could represent the "peaking" volume that is contemporary. You could also chose a volume that represents the inability of the market to "operate" using its rules. Another volume in between could be used to signal to you that failure of the paterns are imminent because volume is not at a "sustaining" level. More commonly used annotations that are very helpful are the ones that are set up for price to fill. Making money with high velocity depends upon using the long diagonal of a trend parallelogram. Why not draw it in in advance so that price can fill up the formation. As long as you are at it, put all the pattrns and formations in place ASAP in the blank space to the right of the forming bar. you will really like the three types of pennants once you catch on to annotaing in the future. Indicators, most people think are lagging. Annotations of indicators are easily done to turn them into leading indicators. This applies to both kinds of indicators. Oscillators are the easier of the two it turns out. One of the worst misconceptions of TA is that the forming bar on the chart is on the very right of the display. Only people with the misconception that annotaing TA is done after the fact. Why would anyone take the time to draw lines on the past?? the only useful pattern and formation anotations are the ones where the end of the formation is composed as far as necessary to the right of the presently forming bar. What good are candesticks? Do they form just after the bar is finished? Did you ever see volume rising and getting to the annotated level for price BO before price BO's? LOL, the BO of price is always after the critical volume is established for the BO. Like someone said: what if everyone had the right third of their screens annotated ahead of price bar formation? They probably think the market would stop following its rules...LOL.....and all the profits would go away... LOL.... PUT a MARK on the MACD where the crossover will occur....really is there anyone out there using the original MACD defaults???...LOL....Do you use as an signal where the lines go from divergence to convergence??? (What does that mean???) Are you still waiting for the cross over to come up. Do you know which line is always horizontal at the time of crossover?? We all know that no one is going to think about any of these comments.......LOL... Why think about them?? I just post something that you have never read before in each post I make in order to let your guys know that you have something (a lot) to look forward to. Everyone neutral biased yet???? Have some eggnog... see if your software lets you move the forming bar off the right side of the chart...LOL.... Who would have screwed that programming function up if you couldn't?