Tendential planimetry method - page 2

 

Vinin, thank you ! Looked at the link, but I still meant a little differently. The author is talking about which gouppies are at the top and which are at the bottom of the indicator. It reminds me of the masterforex fan system, only there are more slices. It is nothing else but a system based on intersection of slices, only supported by other indicators and a bit more reliable. I'm not focusing on the position of the tiles relative to each other, but on their dense areas. I.e. their density covering the chart, but not the upper and the lower one. And I'm not talking about the trend direction (which is actually visible on the price chart alone), but about possible reversal zones and important price levels, which I associate with areas where the bags are particularly dense. For example, the situation looks like this moment (I chose quiet blue coloring to sparing others' psyche :)

(Parameters N=100, Min=2, Step=5. The rest is default)

From the 9th to the 17th of October we see a flat. You can see that it is bounded by a powerful tourniquet from the bottom. We also see that it pulls the lines to itself and then its trace connects with the bunch supporting it from the bottom. Finally, the price decided to go to the sell. On the 17th of October, at 5pm, it broke the strong supporting harness from the bottom and the heavy move started. What stopped it, I don't know. The harnesses don't show that. If they did, they would be the Grail :). At last, on the 20th of October the price went up. But on the 22-th it met the strong bunch from the upper side, so it failed to break through it and went down. On the 23-rd it came to the thin pair from the lower side, but on the 25-th it broke through it and went down. On the 29th a small flat, from which a small tourniquet was pulled. In the meantime, the big bunch from the top significantly dissipated. The way to the buy is blocked by a thin bunch from the flat on the 29th and a more powerful one from the 22nd. My forecast: Predominant price movement downwards. But the situation is not too defined. Important price levels are 1.1620 and 1.1670. It is possible to play on the breakdown of 1.1620. Buy stop at 1.1630. The first target is 1.160. The drawdown at 1.1680. But IMHO it is better to cautiously sell, defending the level of 1.1630 (just above the first harness). It would be better to defend with a long lock, because strong moves are unlikely. However, it would be better to smoke the bamboo until the new strong bundle is formed. Or, if you are impatient, you may use other timeframes for analysis. The latter is ALWAYS highly recommended.

Alas, this is just words for now. But there is a way to translate this into a mechanical trading system. You need to calculate the density of the swabs on each bar as a function of price. Unfortunately, I do not know how to draw such a picture in the terminal. Of course, one can use the idea of hatching like in Ishimoku where vertical lines are drawn. But the trouble is that in contrast to Ishimoku, our straight lines must be colored to hatch areas of different density. It will increase drastically the total number of lines in the figure. In MT4, a line is an object. That is for sure associated with significant overhead. And I'm afraid the indicator showing density may seriously suspend the computer, if not cause the terminal to crash altogether. On the other hand, it's not that much of a problem. Images may be drawn in Matlab as well. It will narrow down the circle of participants. But there's nothing you can do about it. If you want to have a serious conversation with serious people on serious subjects, you'd better master a serious tool :)

 
eugenk:

Vinin, thank you ! Looked at the link, but I still meant a little differently. The author is talking about which gouppies are at the top and which are at the bottom of the indicator. It reminds me of the masterforex fan system, only there are more slices. It is nothing else but a system based on intersection of slices, only supported by other indicators and a bit more reliable. I'm not focusing on the position of the tiles relative to each other, but on their dense areas. I.e. their density covering the chart, but not the upper and the lower one. And I'm not talking about the trend direction (which is actually visible on the price chart alone), but about possible reversal zones and important price levels which I connect with areas where the bags are particularly dense. For example, the situation looks like this moment (I chose quiet blue coloring to sparing others' psyche :)

(Parameters N=100, Min=2, Step=5. The rest is default)

From the 9th to the 17th of October we see a flat. You can see that it is bounded by a powerful tourniquet from the bottom. We also see that it pulls the lines to itself and then its trace connects with the bunch supporting it from the bottom. Finally, the price decided to go to the sell. On the 17th of October, at 5pm, it broke the strong supporting harness from the bottom and the heavy move started. What stopped it, I don't know. The harnesses don't show that. If they did, they would be the Grail :). At last, on the 20th of October the price went up. But on the 22-th it met the strong bunch from the upper side, so it failed to break through it and went down. On the 23-rd it came to the thin pair from the lower side, but on the 25-th it broke through it and went down. On the 29th a small flat was formed, from which a small tourniquet was pulled. In the meantime, the big bunch from the top significantly dissipated. The way to the buy is blocked by a thin bunch from the flat on the 29th and a more powerful one from the 22nd. My forecast: Predominant price movement downwards. But the situation is not too certain. It is recommended to sell cautiously, defending the level of 1.1630 (slightly above the first lock-up). It is better to defend with a lot, because strong moves are unlikely. However, it is better to bamboozle until the new strong bundle is formed. Or, if you are impatient, you may use other timeframes for analysis. The latter is ALWAYS highly recommended.

Alas, this is just words for now. But there is a way to translate this into a mechanical trading system. You need to calculate the density of the flapping on each bar as a function of price. Unfortunately, I do not know how to draw such a picture in the terminal. Of course, one can use the idea of hatching like in Ishimoku where vertical lines are drawn. But the trouble is that in contrast to Ishimoku, our straight lines must be colored to hatch areas of different density. It will increase drastically the total number of lines in the figure. In MT4, a line is an object. That is for sure associated with significant overhead. And I'm afraid the indicator showing density may seriously suspend the computer, if not cause the terminal to crash altogether. On the other hand, it's not that much of a problem. Images may be drawn in Matlab as well. It will narrow down the circle of participants. But there's nothing you can do about it. If you want to have a serious conversation with serious people on serious subjects, you'd better master a serious tool :)


Tried to do something like this, but used the visibility of the masks. Which ones we can see, and which ones are hidden from us. Didn't use density. My last post didn't save with the bug in the terminal for some reason. I'll try to produce it again.

Here it is.

 
Vinin, your computer is glitchy. Alas, I've experienced it myself. The thing is that all this rainbow splendour seriously hampers the terminal. And sometimes the picture just doesn't have time to rearrange itself. I fight it this way. When I want to save something, I simply cut off the network so that no ticks go to the terminal. It helps. Tested by electronics :) Besides, you shouldn't build more than a hundred mashes for a saved picture. Another option is to use some screen grabber instead of "Save drawing" menu item. That should also help, although I haven't tried it.
 
eugenk:
Vinin, your computer is glitchy. Alas, I've experienced it myself. The thing is that all this rainbow splendour seriously hampers the terminal. And sometimes the picture just doesn't have time to rearrange itself. I fight it this way. When I want to save something, I simply cut off the network so that no ticks go to the terminal. It helps. Tested by electronics :) Besides, you shouldn't build more than a hundred mashes for a saved picture. Another option is to use some screen grabber instead of "Save drawing" menu item. That should help too, although I haven't tried it.
I know, I just wrote about the glitch. And I know how to deal with them. Besides, it's just an experiment. And I use a maximum of seven mash-ups. I have to consider the possibilities of the technique.
 
eugenk:

"What stopped the price, I don't know. The harnesses don't show that."
And there should be a seal (like a resistance) on the top TF at this point, so it couldn't penetrate it.

 

Vinin, seven mashes is clearly not enough for this approach...

vaa20003, there was no compaction on the senior frame, I looked it myself, but nothing surprising. The method doesn't show everything. If it showed everything, it would be a grail :) And the serious drawback of this method is that the estimation is given only from the movement on long trends. And we see an empty screen by the movement. And right now the dollar is in a long trend of decline. Moreover, the markets are at historical extremes. So not a good time for this method to work at all...

 
eugenk. I have a different approach. I was just curious at the time.
 

Vinin, pardon my curiosity, but which one ? :)

By the way, here's the development of last night's situation. This time on M30 and with the same parameters (I didn't build a new pattern)

As you can see, the price went from the lower loop to the level of 1.1620, which I called very important last night. By the way, the origin of the level 1.1620 is visible. This is a trace of the flat of the 29th. The market's memory of its past steady state, corrected downwards, by the current market direction. So perhaps some Objective Reality is reflected by this method... Folks, I would be very grateful to all of you if you could try this method to test it in relation to prediction and trading in the real market. Alas, now is not the best time for that, as I said above. But what can you do, "you don't choose your times, you live and die in them"...

That's it. Off to the evening.

 
eugenk. Candlestick analysis. Sometimes I step aside a bit if I find something interesting. But otherwise only the analysis of candlesticks. (I did not use the word 'Japanese').
 
I've never used candles, though... I simply have a rigorous approach to the justification of any method. A method may be as complex as you like, but the underlying postulates should be as simple, clear and unquestionable as Euclid's axioms. In Forex, there is only one such thing. The presence of support/resistance lines and reversal zones. Everything else is bogus :). So I try as much as I can to look for it .... By the way, a couple more words of defense and reasoning. Very often people attach special importance to scales with certain periods. For example with Fibonacci numbers. If we look at the statistics, we'll see what happens with them. The market doesn't like any numbers determined a priori, because it floats and breathes all the time... However there is an element of truth in it. Of course these values have nothing to do with magic numbers, because they are not constant in principle. Although sometimes accidentally they may coincide. Unfortunately it will not be possible to pick up magic periods. For while we will accumulate data for such selection, the magic periods will disappear. The construction considered here solves this problem. But not by selection of periods, but collectively. In fact, we find and use magic periods. The only REAL ones, but not the ones that exist in magical brains of analysts (from the word "anus" :) Otherwise we often have to read: "Attention! The price has crossed the MA55. There will be a collapse, panic and taxation". Now, with our knowledge, we can sincerely laugh at such naive tricks :)
Reason: