Quote from yossy:
Db,
When retracements are spoken off, in cases where these are less than 50% of the preceding swing, the likelihood of trend continuing is taken to be higher than when the retracement retraces almost the entire preceding move. This is inferred to be resulting from the strength of the trend in that buyers are happy to jump in early (in case of Uptrend).
My question is that how does it reconcile with the view that we generally look for a close Higher Low or even Double bottom when looking for a Test which is seen at the start of the reversal of a trend.
Is it just that the logic/dynamics of Test are different so it is best that the S/R level is tested as close as possible. And this is different from retracements in a trend. Or is it something else.
Thanks.
In a word, yes
A "Retracement" is a pullback in a trend, one which enables the trader who ISN'T in to enter the move and which enables the trader who IS in to judge the strength of the trend and even to add to his position. A little-r retracement is used casually as a synonym with a pullback or counter-move. This can be confusing to someone new to the idea, but it is unfortunately unavoidable. Fortunately, the confusion doesn't last long.
A /\ or \/ move, then, may be a retracement but not a Retracement, i.e., price is retracing its steps all the way back to where it started. Thus the "trend" turned out to be a very brief one. This isn't necessarily "weakness" but rather a characteristic of the trading environment which suggests that traders don't know what they want to do and are thus running in place until they decide. A selling wave that is equivalent to a preceding buying wave, or vice-versa, is definitely NOT a signal in and of itself that price is heading off into the opposite direction. It's a sort of "let's try that again" reset.
As for various tests of support and resistance, they of course depend on the existence of support and resistance. If, for example, price is testing support with a double bottom or a higher low, or even a lower low that rejects the lower low position like it touched a hot stove, then one can be reasonably confident of a reversal. However, if there's no identifiable support there, the "reversal" may be nothing more than a retracement in an ongoing downmove, and traders who take every 2B and 123 and Ross Hook and Dunnigan Swing (and so forth) that they think they see end up with a great many counter-trend trades and a great many losses. And a great deal of confusion.
This is not to say that all reversals take place at easily identifiable support and resistance. Sometimes they occur in the middle of nowhere, for no discernible reason. But that doesn't mean that one should take them all on the offchance that they might turn out to be legitimate. The probability of a reversal turning out to be a true one is far greater at S or R. Therefore, the trader who wants to avoid trading counter-trend will avoid trading reversals that may not be reversals at all unless he has damned good reasons for doing so. Trying to catch the reversal just because you missed entering the original trend is not a good reason.
One of the reasons why I came up with the idea of using demand lines and supply lines was to avoid jumping into false reversals. Those who've read Dunningan's One-Way Formula will have seen the difficulties in trying to catch reversals even with clearly-defined criteria. As good as these criteria may seem in theory, the practical reality is that one makes one counter-trend trade after another until the reversal finally comes, at which point one is so far in the hole that even a substantial reversal move may do no more than bring him back to breakeven. A simple D/S line serves as at least a reality check, though nothing is 100%.