Registered User
Joined: 10/7/2004
Posts: 799
Location: Duluth, GA
|
TC offers powerful flexibility for the chart display, making it easy for the user to change the time-duration represented by a given bar (the Bar Length), and the number of bars that are displayed at once (the Zoom Factor). With so many choices available, it's useful to develop a rationale for what choices are MOST appropriate for YOUR style of trading. This article provides a structure for that evaluation.
The starting point is to define what your TYPICAL HOLD TIME is. This will help you set up your screen properly, and assure that the divergence analysis (or other technical studies, for that matter) is giving you results that are appropriate to your trading style. There are two things that you should consider:
1) Choose an appropriate Bar Length
2) Choose an appropriate Zoom Factor
My opinion is that people can "visually digest" the info of about 10-20 bars at a time - that is, when they look at a group of twenty bars, they really "see" what's been happening during that time. Try this out: set your Price chart to Candlesticks (the most visually-rich style), and press Shift-9 to go to the highest Zoom factor ... this shows the 20 most recent bars. "Soak up" the pattern on the screen, look away from the screen, and try to draw it on a piece of paper. Now, go to Zoom 5, which shows the 99 most recent bars. Try to soak up that entire set (same amount of study time) and draw it ... I suspect you will find your accuracy is a lot worse. If you're a real glutton, try it with Zoom 1 (500 bars). It all just sort of fuzzes out.
Of course there is no NEED to memorize all the bars on the screen, but it is certainly USEFUL to have a good feel for how the most RECENT bars have acted, before you get into a trade and during the trade's progression. So, regarding the BAR LENGTH, I suggest that you pick a value that compresses (or expands) your typical hold time to be shown with about 10-20 bars.
That is, if you typically hold a trade for a couple of weeks to a month, then DAILY bars are a good choice. If you hold a trade for 2-4 months (40-80 trading days), then using FIVE days/bar (i.e. aprx a week) might be appropriate. If your hold time typically is only 2-4 days (13-26 hours) then using HOURLY bars will probably work well for you. If you are a "daytrader" who is typically in and out of a position in 2-4 hours (120-240 min), then I'd suggest 10-MIN bars as a good choice. And so on. The idea is to be able to quickly, unconsciously "visually digest" the info that is representative of the time frame you are trading in.
OK, enough about bar length. Now let's consider another issue ... understanding the "lay of the land" that is relevant to your type of trade. There are a lot of reasons why you should ALWAYS consider the historical context of your stock's price and volume action before entering a trade ... and, to a degree, to help you control the position once you are in it. There are three major price-related considerations, IMHO:
1) The typical wiggliness (volatility) of the prices
2) The current momentum (trendiness) of the prices
3) Likely barriers (support and resistance) to price action
A huge amount could be (and has been) written on each of these topics ... let's assume that you have (or will) do that homework some other time. The question we are addressing here is, in order to evaluate these kinds of historical effects, HOW MUCH history is enough ... and can there be TOO MUCH to be useful? I'll offer some suggestions, but you should evaluate this on your own as well.
Prices generally move in "waves". You've picked your Bar Lengths so that one "wave" (which you are trying to capture) is contained in about 10-20 bars. In order to get a decent idea of the "wiggliness" which affects current conditions, I suggest that you should consider what has been happening over a period that is 1x-2x your typical hold time. Since we have decided to adjust bar lengths so that the trade is usually 10-20 bars (call it 15), then that means we should look at the most recent 15-30 bars to evaluate volatility - which is, in essence, the RISK we are taking when we enter the trade and use a trailing stop (again, this is an important subject that's been covered elsewhere). This means our Zoom factor needs to show at least 30 bars on the screen (this is Shift-8 in TC, by the way).
Much is written about "trends" ... one thing is clear ... there is no "universal" way of measuring them. Another important principle is that you need to consider short, medium and long timeframes in order to usefully evaluate the nature and strength of the trend (or lack of it) that your position will be part of. If we assume that ~15 of your displayed bars represents one "ideal" wave in the short-term trend, then conventional wisdom has it that three such waves (plus intermediate pullbacks) would complete a full move. (This is a grossly oversimplified reduction of Elliott wave theory, btw.)
So, if we assume that the intermendiate pullbacks are about 1/3 of the number of bars as the three waves, then a total number of bars in the medium-trend would be 3x(1+1/3) = 4x your typical-hold number of bars. Therefore, we would want to see 4x15 = 60 bars, which is approximately Zoom-6 in TC.
Support and Resistance (S/R) levels are related to peak/valley "pivot points", and to extended "flat" zones where price doesn't go anywhere. (Many would say that Fibonnacci projections and retracements are an important consideration as well - but that is beyond the scope of this article.) The interesting thing about S/R levels is that they can PERSIST for quite a long time. The length of time that they matter is, I think, directly related to how readily visible they are on a chart, to a large number of traders. S/R levels are largely psychological in nature - people think that prices will not go outside those prior boundaries ... so they adjust their trading patterns accordingly, and voila! we have a self-fulfilling prophecy.
So ... how far back can this "persistance" be relied upon? We're really into guesswork here ... but let's presume that our history needs to show at least one full medium-term up-wave and one full med-term down wave in order to see where the "sticky" points are. Further, let's presume that there is about a third of a med-term wave's worth of consolidation/confusion between those two waves. So, our S/R historical window should be about 2.3x the med-term wave length, which using our prior assumptions and settings, resolves to 2.3x60 = 140 bars, which is shown in TC using Zoom-4.
Therefore, we can conclude that if the Bar Length is set so that our typical trade is shown in 10-20 bars, then:
1) Zoom-8 shows us the bars we might use to evaluate volatility-related Risk
2) Zoom-6 shows us the current medium-term momentum/trend of the prices
3) Zoom-4 shows us the Support & Resistance price levels we might encounter
---------------------------
For reference purposes, here are the number of visible bars at various Zoom factors in TC (somewhat rounded off to make it easier to remember):
Zoom9=20, Zoom8=29, Zoom7=44, Zoom6=66, Zoom5=99, Zoom4=148, Zoom3=222, Zoom2=333, Zoom1=500, Zoom1+=750, Zoom1++=1125
---------------------------
Copyright 2005, James D. Dean - All Rights Reserved
|