For Miyagi - Funnel indicator research

In post #5 I was using Median settings for the example. But in real life I am using Hi and Lo. It's my fault you were confused. If "4-Median" is used, those results with the box not exactly lining with the bars would be possible, even now. But let's both assume that Hi/Lo are the settings we are going to use going forward.

Here we go. This is happening right now on G/U. The 2 BB's are overlapping here just as I said. But if you can make it how I said, then this won't happen any more. The 2 BB's would have locked together a few pips ago...

You always come up with original ideas. But I am afraid this is difficult... I probably could add some codes to detect overlap with the other indi's artifact and lock or freeze it, but the part in problem is about two indis becoming one and starting to work another job, since they are actually separate codes. If you are looking for ultimately is a CZ with a specific price pattern in it then it might be a good idea to try to write an algorithm first to spot the CZs.

Miyagi

Do you have a scanner for below conditions?
- higher lows
- long bottom candle tail

Miyagi, please check your PM's...

Forexhard

I think most of my questions can be addressed to this post. The story of this thread seemed to end with the cliffhanger of how 2 or even more separate indicators where going to interact with each other. Is this still the idea. Is it because you want to have multiple shifts. Would it be possible to define this behavior with a single high box def and low box def and then a range of shifts?

How many boxes are you going to have in a real scenario?

Are you going to use this manually? Like in the example you have an uptrend chart and the large shift is on the low box. Would you be manually doing this to match the pattern on the charts? Sorry I am trying to visualize the behavior of these boxes without writing the code yet. It seems like if you also have a high def box with high shift and low def box with low shift then boxes are going to always be joining.

What is the purpose of the center line size? It seems they will lock on the same logic no matter what size they are and then it just disappears.

Can multiple (more than 2 boxes) be locked together.

The picture you have the boxes contain 41 bars not counting the current bar. They are not suppose to expand to the current bar are they? If I am reading you correctly they just lock and stay locked at that time and are still offset boxes.

What is the unlock criteria?

How does the MN Alert algorithm relate to these moving boxes?

How does the heatmap relate? You say it goes in the space between moving boxes. But when they join there is not space right? So it is irrelevant after a join?


Thanks for all you have shared.

For anyone who want to see what boxes that follow price and collide as described in this thread here is my initial stab at visualizing this. I tried to make an animated gif but FF just scrubbed it. So I have included the pictures so you can just scroll through it if you don't want to run a simulation. I think this requires a lot of playing around with to get some useful settings. I can't imagine doing this in realtime so I included my blanksim ea. Just put this in experts folder and save a chart with indicator as blanksim.tpl. Then run strategy tester and away you go.

I have tried to make the initial settings something that is somewhat useful. Basically they attempt to show telescoping consolidation of price action. I am going to continue to try to refine this and any suggestions are very welcome.

If there is an error in the inputs you will see some information displayed in the comments. You can open the file to see what valid values are allowed. All input counts (ie the comma delimited lists need to have exactly the same count).