Quote from Zen Student:
There's the rub. If you know what is going to happen and how to analyse it correctly, you can indeed place consistently profitable and accurate trades in anticipation of this.
What you call "price action" is nothing more than what has happened in the past. This in itself does not give the future unless one can see the decisions which have been made, the reasons for the prior price action, and therefore where the market now needs to go (with a high degree of probability).
The answer is that you can use the market data from the past (price action) to predict what is likely to happen next, but only if you understand the "why" of the prior action. To say that it happened is not good enough. If you compare patterns visually only, you will be comparing things which appear to be the same condition to the untrained eye but which are totally different. This is why you see % success rates quoted for "pattern recognition" - a pattern "fails" z% of the time. If one was comparing like with like, the pattern would resolve in a similar way all of the time.
Your so called "price action" refers to two things: the prior market data, and a rudimentary method for interpreting it which is public knowledge. The limitations of this methodology are well known amongst professional traders...do some even use it to their advantage by setting traps for the unwary?
Lots of questions - your results will depend on how much responsibility you take for your thinking. Success or failure is a matter of thinking or lack of it. Opportunity open to all - happy hunting!
Saying that price action is "only what happened in the past" (although, obviously, the decisions which you find so important also occurred in the past and, as such, can be revisited at any time following their having been made) completely ignores the fact of autocorrelation. Given that we know more autocorrelation exists in the market time series than the efficient market hypothesis would allow for, we can assume that it is possible for informed market participants to identify moments at which the persistence of price direction are more likely to occur. That analysis and identification is best done, I would argue, on the price series itself, not on some external factor to the price series. If a doctor wants to know how a patient is feeling, the doctor doesn't ask one of the patient's relatives (all things being equal, like the patient is able to communicate) the doctor asks the patient. Asking about the "why" of price movements is like asking the relative. Just ask the patient, i.e. the price series. It's actually basic common sense. That it is difficult is no argument against making the attempt.
I already told you the "why" for all price action: market participant disagreement. Knowing the cause for the disagreement is secondary. Why waste time examining secondary causes?
Price action need not be based on "pattern recognition", by the way.