Darn,
You've got my attention again. We are in two of the (imho) silly loops that we keep getting into in trading arguments. Loop 1 is the predict/don't predict argument. Loop 2 is the lagging/leading argument.
To a degree both are rubbish. To a degree both seem to be "more experienced (??)" traders getting frustrated with misconceptions less experienced traders display. In particular the hope that anything will give you certainty, and that more things will give you more certainty.
The way I, personally, display this problem is trying to apply volume, market profile, market delta, volume momentum or anything but price derived data to the HSI. I am searching for a better probability of being right (instead of just trading with the probabilities I get from my method).
So lets attack the misconceptions (and hopefully not make enemies while I do but it won't be the first time if I do):
1) Predict/Don't predict.
predict
v 1: make a prediction about; tell in advance; "Call the outcome
of an election" [syn: foretell, prognosticate, call,
forebode, anticipate, promise]
2:
indicate by signs; "These signs bode bad news"
Everything we do in choosing our
entries and exits is about
predicting. If you buy at the bounce on the LVA you are predicting that there is a high probability that you will move down X before you retrace Y and thus the trade will have a certain win ratio and win/loss that gives a good expectancy.
You are predicting. What you are not doing (I hope) is making the beginners mistake of assuming that there is a certainty associated with the prediction.
Similarly when I see price bounce of a carefully chosen ma after making (say) only 2 thrusts away from the ma I am predicting continuation (x% chance based on history) and thus enter a trade. A second bounce of the same MA may have a higher or lower probability and thus may cause me to increase my bet size or lower it.
Even going with the flow is forecasting. We forecast that if its flowing up now then it will keep going up for Y with a risk of X.
But lets be realistic; if we analyse price and volume, we then predict a probable outcome. Similarly at exit time we exit because the probability of continuation drops below a certain level.
2) Leading vs Lagging.
If you smooth data to build an indicator then the data will have lag built in - and T3 mas, hull mas, and jurik everthings are an attempt to use better signal processing algorithms to get more smoothing for a given time shift in the output (lag).
But. CCI's don't lag the data because the prime determinant of the output is the current price of the current bar.
But. It doesn't really matter anyway for many forms of discretionary trading whether there is a lag in the indicator or not. Its whether there is a lag in the usage of the indicator. For example, EMAs have plenty of lag. But I don't care because I use an ema the same way you use a
POC ... its support and resistance and
as support and resistance it has no lag. The ema represents in real time with no lag exactly what every other trader looking at it as support and resistance sees so if price reacts appropriately at the ema (or the cci does) then I react without lag to reality.
But. Pure price can have lag. If you buy support then thats "zero lag." If you wait for it to reject support thats lag. If you wait for it to form a double bottom at support thats more lag (vs just buying). So even in pure price trading you have lag and the lag is a trade off of increased certainly for time and probably position.
So IMO predicting is what we do. What we must understand is that every prediction has a probability of being wrong so we must have an exit when its wrong and we must
really understand that its probabilistic. The enemy is false confidence (and resulting disillusionment) not prediction.
And how bad is lag. This is a game of predicting and probabilities and risk vs reward. We trade two types of risk: risk of being wrong vs risk that when we're wrong we give more back. What do I mean by that? If you buy at the ema with a 3 pt stop you might have a 50% chance of being wrong but only risk 3pts to find out. If you buy after price forms a rejection pattern from the ema you may need to use a 7 point stop but have only a 30% chance that you are wrong. The second had lag but in return for the lag you get a higher probability of being right (at the expense of a higher risk and less reward (because you are part of the way to your target)).
OK. Monologue over. Hopefully that makes sense.