Iterative Refinement

Following this post I will spend the rest of the weekend attempting to forget today's trading effort.

Since my previous analysis concerning the 10:50 to 11:55 period was incorrect then I can only offer the following:

There was no signal of change from a short position given the entire day. One needed to determine that the first bar of the day was a signal to short and remain that way for the entire day or at least until 15:25.

Time for a nap. Good weekend to all.
 
David,

I am scratching my head after reading your post and how you reached your conclusions.

Quote from dkm:
11:45 fbp bo + ibv We now have pt3, but it is also a pt 1 as we are following a trav level sfc
More...
If you had a point 1 at 11:00am, that must have been a traverse level signal for change and 11:40 would be your point 3. So how can you say that 11:40 is your traverse level change.


Quote from dkm:
*11:55 jw but we are on the way to a pt2 so invalid sfc
More...
Until about 12:00/12:10 with 12:30 confirming it, we did not know we were possibly building another traverse.

For me the pennant FBO at 11:00 was traverse level change and my point 1. Point 3 was at 11:40. It was not until 12:30 did we have a traverse that morphed into a channel.

As for why 11:55 was not change. I am not quite sure the reason, but something about the way the bar closed told me we were not done.
 
Quote from treeline:

So perhaps increasing Price on decreasing Volume and increasing volatility is not a Traverse-level change signal.
More...

I was thinking along the same lines as well after looking at the bars. However there is another difference between the two bars I noticed in real time as the bars were forming that I will add to the discussion.

11:55 started as a flaw bar with really low prv and the prv picked up at the end of the bar (ie if the bar closed 10secs later we might have got increasing black volume). In contrast 12:35 started with increasing prv & about equal volume (prv wise to the prior bar) and continually decreased through the course of the 5 minutes.
 
Just wanted to get this posted from today.

Odd day. Considering the news of the past week. Big events took place.

This is what I had by the days end.
Again. . .
Just my view.

On my way out.
Have a great weekend everyone.

ES 5 minute September 19 08.
 

Attachments

Quote from bi9foot:

David,

I am scratching my head after reading your post and how you reached your conclusions.

If you had a point 1 at 11:00am, that must have been a traverse level signal for change and 11:40 would be your point 3. So how can you say that 11:40 is your traverse level change.

Until about 12:00/12:10 with 12:30 confirming it, we did not know we were possibly building another traverse.

For me the pennant FBO at 11:00 was traverse level change and my point 1. Point 3 was at 11:40. It was not until 12:30 did we have a traverse that morphed into a channel.

As for why 11:55 was not change. I am not quite sure the reason, but something about the way the bar closed told me we were not done.
More...

Firstly, this conclusion was only reached with hindsight. Secondly, after further consideration, I think it is wrong. :)

The difference between 11:55 and 12:35 is that 11:55 is before a pt2 and 12:35 is after a pt3. This requires us to restart the 1,2,3 sequence at or after the arrival of the pt3 at 11:40. The problem boils down to knowing when a "sequence repeat" begins and I am still unclear when it is valid to do so.

Since my last post, I have noticed that the fbp BO at 11:45 was confirmed by the 11:50 bar. This MAY be sufficient reason for "sequence repeat" - I am not sure. But after the lat bo at 12:30 we did NOT get more ibv so this is definitely a difference.
 
Quote from bi9foot:

11:55 started as a flaw bar with really low prv and the prv picked up at the end of the bar (ie if the bar closed 10secs later we might have got increasing black volume). In contrast 12:35 started with increasing prv & about equal volume (prv wise to the prior bar) and continually decreased through the course of the 5 minutes.
More...
Thanks for this info; I wasn't monitoring in real time today. Based on your observations, it makes perfect sense that the 11:55 bar would show increasing volatility, and that the 12:35 bar would show decreasing volatility. It seems that the 11:55 bar was an instance when our 5 minute bars weren't quite able to pinpoint the shift from non-dominant movement back to dominant movement.
 
Quote from ehorn:

Peak Volume OR Pace Acceleration That is the question...

I was hoping to stimulate some more discussion regarding these concepts. Also, how does "Volume Surges" fit into these concepts. I often observe Peak Volume occur as an LTL end effect (or nearing) of an L2 traverse and/or L3 channel and Pace Acceleration occur during BO of L2, L3 channels. I also observe volume surges during turning points within L2 traverses (for instance a bar that slams up, creating an opening bar for LM retrace). Can anyone else comment on their observations and/or findings?
More...
For a while I thought that Peak Volume could be differentiated from Pace Acceleration by looking at the relative Volume peaks within the current Traverse. I thought that Peak Volume was never the bar with the highest Volume of the current Traverse. This seemed to make sense based on the Jokari Window. If we were to zoom out to a slower fractal, our Traverse would turn into a Tape and the Jokari Window says we should see Volume decreasing during dominant movement as we near the end of that Tape. We don't zoom out to slower fractals, but we should be able to see this decreasing dominant Volume on the 5 minute Traverse fractal, in the form of decreasing Volume peaks. Most of the time, Peak Volume does have a lower peak than at least one other bar in the Traverse, but I've now seen 2 instances when Spydertrader confirmed a Peak Volume bar which had the highest Volume in the Traverse. These bars were at 10:55 EST on 9/15/08 and at 11:35 EST on 8/1/08.

So I'm back to the drawing board. Right now I'm considering the possibility that any acceleration of the Gaussian slope that appears after all sequences have been completed should be viewed as Peak Volume in real time. By this I mean not only that P3 must have already been confirmed, but also that any Lateral Breakouts (even those occurring after P3) must have already been confirmed. In other words, the bar that confirms a P3 cannot also be Peak Volume, and the bar that confirms a Lateral BO with more increasing Volume cannot also be Peak Volume. I need to see a lot of charts before I come to any conclusions about this hypothesis, but it fits well with your observations that Peak Volume tends to occur late in the trend, near the LTL, whereas Pace Acceleration tends to occur upon BO of the previous trend's RTL.

I'll consider your thoughts on "Volume Surges" as I monitor over the next several days.
 
Quote from guavaman:

There was no signal of change from a short position given the entire day. One needed to determine that the first bar of the day was a signal to short and remain that way for the entire day or at least until 15:25.
More...

Incorrect. Try again.

Two change signals from short to long exist. Both develop in the final sequence of a channel. Most likely, you've annotated them in such a way that you do not 'see' them as you should. Start with Volume (since Volume leads Price) and look where these changes might be. Follow the market mode in the same fashion as I previously described, where currently, you feel confusion exists.

The market can provide the clarity you require.

- Spydertrader
 
You must log in or register to reply here.
Back
Top