SPY Daily Composite

[gassah 10]

The attached shows the daily composites for the two ranges using Market Analyst. The blue zones are value and the green and red are the 2nd and 3rd standard deviations, respectively. Barros likes to look at the red zones as the extremes of the ranges and it's interesting how they overlap for the top and lower ranges.

 

If price gets turned down from the extreme of the lower range then it might represent a good shorting opportunity. If it can return above the pink line with conviction, back into the upper range, then the old highs as a target seem probable, imo.

 

nic

[DbPhoenix 42]

This in essence is what I've been doing to find support and resistance, though I've been using plain ol' volume at price and eyeballs. The utility of it is that price will reverse at extremes and return to the "value area" (or POC, or midpoint, or whatever), i.e., that "big money" traders will find few trades at the extremes and return to those areas where they will be able to do business. For anyone looking to ride the wave, they generally begin at those extremes.

 

 

Good to see you back

[gassah 10]

Good to see you back

 

Thanks. I'm lovin' MP. I haven't been this turned on by TA since you introduced me to Wyckoff eight years ago!

 

nic

[DbPhoenix 42]

I did? Apparently some of my past has leaked out in order to make room for new stuff.

[heretodaygone... 10]

gassah,

 

How does standard deviations help you with your trading? I can possibly see them as profittaking areas but haven't put any time into their usefullness.

[gassah 10]

gassah,

 

How does standard deviations help you with your trading? I can possibly see them as profittaking areas but haven't put any time into their usefullness.

 

They are just one way to delineate the extremes of a range, to take profits or to initiate a position.

 

nic

[heretodaygone... 10]

Ok, after looking some more, I can see what your looking at. Would you happen to have an intraday chart of possibly a few days worth of data?

 

thanks

[gassah 10]

Ok, after looking some more, I can see what your looking at. Would you happen to have an intraday chart of possibly a few days worth of data?

 

thanks

 

Sorry, I'm only setup in Market Analyst with Forex for intraday data.

 

nic

[heretodaygone... 10]

thanks anyway...good stuff

[DbPhoenix 42]

They are just one way to delineate the extremes of a range, to take profits or to initiate a position.

 

nic

 

I'm not a student of MP, but I understand that it considers the value area to include 70% of the trading activity. Since the center portion of the bell curve represents 68%, I assume that this is what they have in mind. Given that, the tails on either side would represent the same as the bell curve, 13% and 2% on one end and the same on the other.

 

I learned the same support and resistance dogma as everybody else, and it made sense, as far as it went, so that's what I've worked with all these years. But auction market theory fleshed it out and deepened my understanding of just what is going on with regard to S/R.

 

Someone wrote that if you combine the principles of Wyckoff with the principles of auction market theory, you have about as close to a grail as you're going to get. He may be right.

[heretodaygone... 10]

I'm not a student of MP, but I understand that it considers the value area to include 70% of the trading activity. Since the center portion of the bell curve represents 68%, I assume that this is what they have in mind. Given that, the tails on either side would represent the same as the bell curve, 13% and 2% on one end and the same on the other.

 

I learned the same support and resistance dogma as everybody else, and it made sense, as far as it went, so that's what I've worked with all these years. But auction market theory fleshed it out and deepened my understanding of just what is going on with regard to S/R.

 

Someone wrote that if you combine the principles of Wyckoff with the principles of auction market theory, you have about as close to a grail as you're going to get. He may be right.

 

I don't know if I am asking this question correctly, but, if 68% is the center portion of the bell curve and the POC doesn't necessarily have to be at 68%, then how would (could) one calculate range extension(standard deviation)? Is it calculated from the bell curve median or the POC?

[DbPhoenix 42]

As I said, I'm not a student of MP. But there are three measures of central tendency, and if the "value area" is constructed about the number of trades that take place at a given price, I'm assuming that these measures will be so close that any differences won't matter.

 

As far as range extensions, I believe that they are simply movements beyond the range, but there may be more to it than that. When they occur, of course, the distribution will no longer be symmetrical. It will be skewed. When this occurs, the value area has likely shifted, as it did in the NQ on Friday. Whether it stays there or not remains to be seen on Monday.

 

There are plenty of MP devotees on this site. Shouldn't be hard to find more authoritative answers than mine.

[heretodaygone... 10]

and if the "value area" is constructed about the number of trades that take place at a given price, I'm assuming that these measures will be so close that any differences won't matter.

 

 

actually a lot of times they are not which gives the trader asymetrical opportunities(thats a good thing:))...thus the question I asked. It may be significant to let a trade run more if one knew where to calculate from...very interesting.

[DbPhoenix 42]

If they're not, then the distribution won't be symmetrical. Your question was based on a symmetrical distribution, a bell curve.

 

As to letting the trade run, if one receives no reversal signal at the extremes, then he lets it run. No reason to exit for no other reason than arriving at a number.

[heretodaygone... 10]

arghh,you've been doing this too long:)...go with the flow. Here's a mp chart to show what I mean.

[heretodaygone... 10]

so, as you can see friday shows a symetrical distribution which then an easily definable standard deviation(range extension ) can be calculated. BUT, Wednesday and Thursday had asymetrical distributions. So the question remains, how does one logically determine standard deviation(range extension ) ?

[DbPhoenix 42]

As I said, you'll have to ask an MP person. Sorry.

[heretodaygone... 10]

the question wasn't meant for you per say...gassah chart has me thinking. I thought you might have run across this before. thanks

[gassah 10]

The Market Analyst software isn't required for the standard deviation calculations. Barros states that the 12.5/87.5% fibonaccis make for a good approximation of 3rd STD, as does the 33/66% for value (1st STD).

 

nic

[gassah 10]

The previous post should have stated 33/67% (not 66%).

 

From the CBOT manual:

 

In order to understand how the distribution process relates to market activity, it's important to see the connection between the

Market Profile concept and volume. The volume of everything

typically falls one, two or three standard deviations from the mean. We're going to relate trading data to this organization. For our purposes, however, we're just going to relate the high volume first standard deviation and the low volume third standard deviation to

market activity.

 

The first standard deviation correlates to the value area. This is a high volume area. It shows price acceptance confirmed by use: a fair price area.

 

The third standard deviation correlates to a price excess. This is a low volume area. It shows price rejection: an unfair price area.

 

These low volume price areas are key reference points because they can contain the range. When the market reaches these potential parameters, it can only do one of two things: trade through or reverse direction.

 

nic

[OAC 10]

The Market Analyst software isn't required for the standard deviation calculations. Barros states that the 12.5/87.5% fibonaccis make for a good approximation of 3rd STD, as does the 33/66% for value (1st STD).

 

nic

 

nic,

Where do you get the Fib Ratio of 87.5% ? The closest ones I know are 78.6% and 88.6%.

One would arrive at 78.6% by taking the reciprical of 61.8% which is 161.8%, taking its Square Root equals 127.2%, then take the reciprical again equals 78.6%.

And 88.6% is simply the Square Root of 78.6%.

[gassah 10]

nic,

Where do you get the Fib Ratio of 87.5% ? The closest ones I know are 78.6% and 88.6%.

One would arrive at 78.6% by taking the reciprical of 61.8% which is 161.8%, taking its Square Root equals 127.2%, then take the reciprical again equals 78.6%.

And 88.6% is simply the Square Root of 78.6%.

 

100-12.5= 87.5

 

nic

[OAC 10]

By the way for those who never seen it, this is where all the Fib numbers came from:

Fibs.doc

[DbPhoenix 42]

Of course, one must be careful not to focus so much on the form that he overlooks the substance:

 

The first standard deviation correlates to the value area. This is a high volume area. It shows price acceptance confirmed by use: a fair price area.

 

The third standard deviation correlates to a price excess. This is a low volume area. It shows price rejection: an unfair price area.

 

These low volume price areas are key reference points because they can contain the range. When the market reaches these potential parameters, it can only do one of two things: trade through or reverse direction.

Edited April 6, 2008 by DbPhoenix

[BlowFish 10]

Gassah, I guess you have read JPerls's excellent series of threads elsewhere on this site? His approach is based on volume profiles rather than TPO's (which I'm pretty sure was adopted by Steidlmayer in the early days as a proxy for volume). Jerry's slant really resonated and its probably as well structured and written series of threads that you will find.