Potential NQ Targets
- Thread starter dbphoenix
- Start date
I should point out here that the integrity of this approach has absolutely nothing to do with how -- or whether -- it is implemented. No matter who is trading how, price retraced 10m after the opening plunge. Whether or not the trader took the trade has nothing to do with the fact of the retracement.
Why this retracement rather than one of the many others which occurred beforehand? Because we're at the upper limit of the trend channel. Without context, one is just stumbling around in the dark with a penlite.
Why this retracement rather than one of the many others which occurred beforehand? Because we're at the upper limit of the trend channel. Without context, one is just stumbling around in the dark with a penlite.
This is what I struggled with in terms of backtesting. In some respects, backtesting seems a bit "hollow" in that you're just looking at statistical probabilities based on as few variables as possible so as not to over complicate I think. Each time you introduce a new variable, you need to further segment the data.
As I go through my charts and look at retracements, if I filter out which of these are close to hourly or daily support/resistance, and where this occurs in the trend channel, there might perhaps be a way to increase the win rate. But this amount of work would almost be impossible and difficult to track because you'd have to seperate out all the retracements that happen in some context and hence there might be many different "types" of retracements. (ie. RETs close to trend channel extremes.... RETs close to the mean.... RETs close to PDHs and PDLs)
(As an aside, the important RETs that I want to study next after I'm done with 5 min line breaks is RETs after big drops or rises.... but once again, this won't have the context of the day/week built in, just simply the context of a price bar being much larger than the other bars around and hence a quick drop or rise)
Of course as you say though, given context, and knowing the LOLR, this makes trading in that direction that much more likely, and hence makes that retracement that much more likely to produce a very good move. So what I'm left with I think is still that you gotta backtest certain strategies, but always be left thinking a bit in real time and having a feel for when to apply a certain strategy. (ie. perhaps making sure to be selective with whatever longs might set up on a day like today even though an initial line break followed by a RET might occur. Since we never know what will happen, cherry picking trades is dangerous, but if the LOLR is down given the channel top, is it fair to skip some longs that might set up mechanically?)
Are there any glaring errors in this line of thought?
Whether or not there are "errors" depends on what your goals are and what you're developing in order to reach those goals. In terms of the SLA, one always begins with the weekly and daily charts, i.e., context. If one hasn't begun there, then whether he uses a 1m or 5m or 60m bar is irrelevant. He is instead trading a different approach, perhaps even scalping.
This is what I'm realizing as well and I'm not quite happy with it. But proving to myself to not be scared of the retracement is a critical first step I think. Learning to hold a trade is the next step, but just being able to get in, and at the same time knowing what I'm risking by having an idea of where the stop would be and accepting that risk will hopefully help this process along. Re-entry is always possible, but just like today, when the trade works, it just works.
The heavily overlapping bars after that first RET entry sure doesn't help, but seeing price is unable to rise, and the fact that it stays below the 50% level of the first drop I guess are all good clues to help keep you in the trade.
It really is amazing how channel boundaries function, whether the mechanism is mean reversion or mass observation of the channel or some mixture. In the end it just seems to be an expression of human fear and greed at work. Very interesting area now, as always.