Custom Search

Theoretical support/resistance (S/R) level

In the preceding article we identified the theoretical support/resistance (S/R) level with the volume-weighted average price at the   which the stock or commodity had changed hands during an as-yet-unspecified interval of time. In the same spirit of logical   development, we now motivate the choice for this interval.

Consider the familiar case of a stock which has been dormant for a long period of time, trading in a narrow price band on   relatively low volume. On a certain day it suddenly breaks out of the trading range on heavy volume and we ask where we might   expect to find support during the inevitable pullback which follows when the buying spurt subsides. We have already said that the   theoretical support will involve an averaging process from some initial instant (called the "launch point") to the present.

If the launch point includes days prior to the sudden breakout, the averaging will mix time periods of differing underlying   psychology and thus would not be expected to yield useful results. For, clearly, the breakout day marked the beginning of a shift in   the psychology of those subsequently acquiring the stock. In other words by and large people were buying the stock for different   reasons after the breakout than before. Similarly we can consider another familiar situtation where a stock has been in a long   period of consolidation after an earlier bull move. Again, volume has shrunk to a mere shadow of previous levels.

Then, one day, the stock starts moving up and trading volume accelerates. Here again the trend reversal is indicative of a reversal   in underlyling psychology - else why would there have been a change in trend? Thus it is clear that if the average price is to be a   meaningful measure of psychological boundary, the average must be taken over a period of homogeneous psychology, i.e.   subsequent to a reversal of trend. This is the key to the Midas method. By one of those unfortunate detours in the progress of   technical analysis, attention came to be focussed on moving averages, i.e. price averages taken over a fixed time interval from the   present to the past, an interval which had no connection to the underlying psychology of the market participants (except perhaps   for a 6-month capital gains holding period).   Our "message" is that instead of "moving" averages, one should take fixed or   "anchored" averages, where the anchoring point is the point of trend reversal.

By way of example, consider the Midas chart (obv omitted to focus on price alone) for Cypress Semiconductor. Here we have a   five-fold hierarchy of theoretical support levels, where every member of the hierarchy is launched precisely from a trend reversal   point. Thus an otherwise bewilderingly complex set of zig-zags in the price history can be understood with respect to a single   algorithmic prescription: support will be found at the volume-weighted average price taken over an interval subsequent to a reversal in trend.

"Is this always the case?", one should now ask. In Count Dracula's immortal words when his hapless ovenight guests wished to   leave: "Ahh, if only life were so simple!" Have a look at the Midas chart for Cycare Systems. Here we could achieve a good "fit"   to the observed price zig-zags only by adjusting the launch points somewhat from their a priori values. In the case of S2, is was a   matter of determining where in the midst of an extended consolidation bottom the launch point should taken. With S3 and S4, it   involved a displacement of a day or two after the actual reversal day (a common situation in strongly trending stocks). (S5 could   be viewed actually as an S6 - launched from the first pullback to an S5 which is not shown.)

In all cases, we attempt to understand the actual price zig-zags within a framework of the minimum number of theoretical S/R's.   The launch points for this minimal S/R hierarchy are in the final analysis empirically determined. However, in most cases, the   initial guess of launching the S/R's from the trend reversal points is good enough.

Having thus developed the rationale for the Midas method of technical analysis, we will henceforth get down to the computational   nitty gritty. After the next article (if not by now) you should be generating your own Midas charts and forming your own conclusions!