The Price / Volume Relationship

[innersky 10]

As the gaussians more or less do what I want them to do, this is still not anywhere near something that actually works. It struck me that I also need price. And more specifically, VE's. No kidding Sherlock :-).

So next step: automated channel drawing.

 

--

innersky

[gltrader 10]

Hi.

 

This is my chart for 11/05/13

 

any comments are welcome.

[plantrader 10]

Hi there all fractal traders. I've spent a few hours reading this thread, although admittedly haven't taught myself how to do it yet. I find it very interesting, though, but since the only question I can ask right now is a general one... would anyone mind? Based on some of what I've read...

 

1) Due to the nature of this being both quantitative rules-based, but also a bit of an art (or abstract), do you find that your mistakes are mainly recognizable in hindsight, over and over and over, due to the aspect of subjectivity involved which can't be accounted for upfront in a 'rule'? Or can you be correct enough of the time so that that aspect isn't a major concern? Just curious. I'm not saying this is necessarily an issue, but it would help to get a perspective on that.

 

2) Has anyone started from scratch learning this, based on this thread, and taught him/herself sufficiently enough savvy with this to trade consistently profitable using it? I mean, other than the teacher SpyderTrader.

 

I know I'm new here, and my questions are general... please don't take offense. Part of how some of us learn is to go through a process of analysis and elimination of the initial questions which pop up (so that they don't become mental roadblocks to learning).

 

p.s. I like the way SpyderTrader's replies have a Zen aspect to them. It reminded me of a movie where someone is training to be a Ninja. Will NinjaTrader suffice? ;-)

[plantrader 10]

Hi again. I've got a few more specific questions. I'm going to have to dig, practice and repeat until this becomes second-nature. But here are some questions I didn't get clarity on in the thread yet.

 

1) On page 17 of the thread, the acronym "FTT" is used but wasn't previously defined except for in the first page where it's shown in a screenshot/diagram. From the diagram, it looks like it's a first-time-touch to a new trend (traverse) when price changes direction? Or is it moreso the last data point when a trend ends? If they aren't always one in the same (end/start of trend).

 

2) Does B2B mean "black-to-black"? Does "2R" mean "to-red"? And "2B" means "to-black"? Or does the usage of the number 2 also indicate the # or count of volume bars of the same color? Such as needing 2 black volume bars in sequence or if not in sequence maybe 2 black bars of increasing volume (with respect to one another) in order to know that it's correct in drawing a black line via hiking up the mountain?

 

3) Do the Gaussian volume patterns always appear in the same exact sequence? Either B2B 2R 2B, or R2R 2B 2R, and nothing else? Or do we get valid gaussian sequences which differ from that, and it's our job to identify the ones that do match either of those two sequences if/when they finally do match those sequences (if not all the time)?

 

4) If the gaussians do always match either a) b2b 2r 2b or b) r2r 2b 2r, then can they overlap? Or are they distinct separate units as in a & b, exactly in that order without overlap?

 

5) If the gaussians don't overlap, then in order to constantly repeat either of those sequences over & over without other variations... do we have to change magnitude in order to see them? Such as, we get a b2b 2r 2b on a 5 minute chart, but then in order to see the next sequence (either b2b 2r 2b or r2r 2b 2r) might we have to also be looking at a 3 minute chart because it wouldn't be definable on a 5min chart and there'd be a period of non-valid gaussian sequence on the 5min? Just an example, hopefully you get what I'm asking on this.

 

6) It appears in most (or all?) the examples I've found, after price trendline touch #3 there's a price continuation in the same direction as the current traverse? Meaning, if the guassians & price action fit the 1, 2, 3 pattern... then it seems somewhat reliable that after #3 there will be a price move in the expected direction. As a result of the power of the observed volume pattern and observed price action in the current traverse. Possible trade entry signal after data point #3. Since, if my observation is correct, and if it's true (as stated in the thread) that the same fractals/patterns will occur on every scale (small or large re: tapes/traverses/channel) no matter what. Does it follow that the same above rationale could be used for data point 1 or 2 as well, as long as we're savvy enough at identifying and correlating the volume gaussians with the price action?

 

7) Does proper identification of the volume patterns (gaussians) and price action (tapes/traverse/channel) occur in real-time, reliably? Or is it ever-changing with so many possibilities and permutations of, that clarity is only ever gained in hindsight?

 

I hope this isn't too many questions. The best way for me to practice and teach myself to view the market this way (properly) is to have a good enough basis so that I can avoid teaching myself something invalid without ever knowing it (as I practice). The initial basis for understanding is one of the most important building blocks to advancing any learning, since the definitions and knowledge are already a known commodity. And it's what allows the savvy to be developed as we practice and retrain our brain re: identification/differentiation.

 

Thanks for any help and clues, all! Have any of you become comfortable/savvy enough to use this viewpoint/approach/analysis as your primary tool for your trading efforts?

 

Matt

[ptunic 10]

Hey Matt, welcome aboard! I took a break from this years ago, (might one day try again!) but still subscribed to this thread. I can try to answer a few of these.

 

 

 

1) On page 17 of the thread, the acronym "FTT" is used but wasn't previously defined except for in the first page where it's shown in a screenshot/diagram. From the diagram, it looks like it's a first-time-touch to a new trend (traverse) when price changes direction? Or is it moreso the last data point when a trend ends? If they aren't always one in the same (end/start of trend).

 

FTT stands for Failure to Traverse. The basic idea is you have 3 legs consisting of dominant movement, nondominant movement, and then more dominant movement. Also, the 3 inflection points here create a channel. After you have established the 2nd leg of dominant movement (which occurs immediately beginning on the "point 3"), the price will be making a "run" for the channel boundary.

 

The basic idea is if it is able to reach that channel boundary (and with sufficient volume.. this part is somewhat fuzzy to me) -- then the Trend is intact and what will occur next is nondominant movement and then further dominant movement. In other words you will have a "point 4" and "point 5". This can continue indefinitely. There have been entire days where this goes on the whole day, I think I've seen as many as 11 points or something though hypothetically it can go on forever.

 

That said, most of the time you only get 3 or 5 points and then you get the FTT. The FTT occurs when the dominant price action is not sufficient to touch the channel boundary. This is when the trend has changed to another direction. The point 4 of the trend you were on, has now become Point 1 of a new trend.

 

 

 

 

2) Does B2B mean "black-to-black"? Does "2R" mean "to-red"? And "2B" means "to-black"? Or does the usage of the number 2 also indicate the # or count of volume bars of the same color? Such as needing 2 black volume bars in sequence or if not in sequence maybe 2 black bars of increasing volume (with respect to one another) in order to know that it's correct in drawing a black line via hiking up the mountain?

 

Yes, it means black-to-black or red-to-red. The B2B always means black (upward) price movement, first with falling volume, then rising volume. So in both parts of the B2B, price is more or less moving higher (it is "black", aka "upward" price action, after all). The different is in the first part of the B2B, volume is decreasing. This (almost?) always happens right after the FTT, while you are still overlapping the previous channel.

 

Then, the 2nd part of B2B occurs when price has exited the previous (red) channel. You (almost?) always see a big boost in volume at exactly the point where the new channel has crossed the previous channel.

 

 

3) Do the Gaussian volume patterns always appear in the same exact sequence? Either B2B 2R 2B, or R2R 2B 2R, and nothing else? Or do we get valid gaussian sequences which differ from that, and it's our job to identify the ones that do match either of those two sequences if/when they finally do match those sequences (if not all the time)?

 

Well, as I mentioned in 1), trends can move onwards for indefinitely.

 

So, for black channels, you have these combinations:

 

B2B-2R-2B

B2B-2R-2B-2R-2B

B2B-2R-2B-2R-2B-2R-2B

B2B-2R-2B-2R-2B-2R-2B-2R-2B

... etc...

 

 

 

4) If the gaussians do always match either a) b2b 2r 2b or b) r2r 2b 2r, then can they overlap? Or are they distinct separate units as in a & b, exactly in that order without overlap?

 

Well, you have bigger and smaller channels. 3 (or more) small channels together give you 3 (or more) points, which is enough to make a larger channel on a bigger scale. They are fractal in nature, in other words those larger channels in turn can be used to create even bigger ones. So the bigger channels contain smaller channels within them.

 

But, if you are sticking to channels the same "size", then the b2b 2r 2b r2r 2b 2r are distinct, except that the FTT of one becomes the Point 1 of the next.

 

You can think of them more like this:

b2b-2r-2b-FTT -> r2r-2b-2r-FTT etc

 

The FTT and the first r of the r2r are exactly the same Point (specifically Point 1). The channels overlap from the FTT of the previous channel until volume starts to increase in the opposite direction.

 

 

 

 

 

5) If the gaussians don't overlap, then in order to constantly repeat either of those sequences over & over without other variations... do we have to change magnitude in order to see them? Such as, we get a b2b 2r 2b on a 5 minute chart, but then in order to see the next sequence (either b2b 2r 2b or r2r 2b 2r) might we have to also be looking at a 3 minute chart because it wouldn't be definable on a 5min chart and there'd be a period of non-valid gaussian sequence on the 5min? Just an example, hopefully you get what I'm asking on this.

 

I sort of answered this above, but one thing to clarify is it isn't the 3 min or 5 min charts per se that you use to "magnify". Channels are always made up of smaller channels.

 

Also one thing is I would highly recommend sticking to purely the ES 5 minute chart, as that is what 99% of the examples are in. If you gain skill in that, you can always move to other time frames or instruments.. the logic is universal and works on anything (where there is sufficient liquidity, so not penny stocks).

 

6) It appears in most (or all?) the examples I've found, after price trendline touch #3 there's a price continuation in the same direction as the current traverse? Meaning, if the guassians & price action fit the 1, 2, 3 pattern... then it seems somewhat reliable that after #3 there will be a price move in the expected direction. As a result of the power of the observed volume pattern and observed price action in the current traverse. Possible trade entry signal after data point #3. Since, if my observation is correct, and if it's true (as stated in the thread) that the same fractals/patterns will occur on every scale (small or large re: tapes/traverses/channel) no matter what. Does it follow that the same above rationale could be used for data point 1 or 2 as well, as long as we're savvy enough at identifying and correlating the volume gaussians with the price action?

 

Indeed, you have found the most recommended, "safe" entry point for beginners! Find a point 1, point 2, point 3, and as early as you can enter in the direction of that dominant trend. Keep in mind, there certainly are many times when price will move against you (as the FTT can come 1 bar after the point 3, leading to an immediate loss). But, the idea is if you enter immediately after a point 3, your maximum loss is no more than that point 3 bar. Even an FTT inside the channel may still lead to some profit, and if you get a point 4 then that could be a lot of profit.

 

If you get consistent at entering after the Point 3, then the next trade to learn as I understand it is the FTT. This one leads to much higher profits, but also takes much higher skill. Though really every Point is just an FTT at a smaller channel size, but at least of the channel size we typically use (5-30 bars of 5 minutes ES), that would be the next hardest trade to learn (the FTT off that sized channel). Then typically (as I understand it) you don't want to take trades between Point 2 and Point 3 at least of that time frame, as even if you identify them correctly there just isn't much room there so your risk to reward ratio is too high.

 

 

This said-- I think the general recommendation is don't even try to practice trade without really first annotating charts for a while. Eventually you want to do it of course (trading is the whole goal after all), but there is a lot that can first be learned just by annotating the charts.

 

7) Does proper identification of the volume patterns (gaussians) and price action (tapes/traverse/channel) occur in real-time, reliably? Or is it ever-changing with so many possibilities and permutations of, that clarity is only ever gained in hindsight?

That depends on your skill level. At a certain level of mastery, you are able to consistently anticipate where things are headed, allowing you to profitably trade via the method. I wasn't able to get to that point, but several of the students here who I met in person have been trading this method full time as their only source of income for years. It is by no means easy to learn, I'll admit (to my knowledge there are just a few who have), but I believe it can be done by anyone with time and dedication.

[plantrader 10]

Thanks a lot for the thoughtful replies, ptunic! I appreciate that. I'll ask some follow-up questions as below, for whevenever/if you get some time to share again. Did you take a break from this because you were having too hard a time defining (differentiation) things clearly enough to make it actionable?

 

1) FTT = failure to traverse; thanks for the clarification. So then, in a vanilla example we'd get a point 3 and then a FTT afterwards at some point. But you mentioned a point 5, and so on. Does that means the 1-2-3 has started over? Or is it a 1-2-3-4-5-6-n sequence when the trend continues? If so, I thought Spyder said that the fractals or sequence was always the same, without exception? I'm probably just not understanding exactly what was meant. Your clarifying this helped me (thanks) "b2b-2r-2b-FTT -> r2r-2b-2r-FTT etc"

 

2) For the Gaussians, and the sequence of either B2B 2R 2B or R2R 2B 2R, is the repeating pattern (or fractal) the simple fact that there will be a channel or traverse with a series of touches? Even if more than 3 (or 4) touches, in cases where the trend continues in the dominant direction overall for most of a day or more than 1 day?

 

2.a) Is there a way to see B2B 2R 2B 2R 2B 2R as something else, and be correct in seeing it? Or is it as simple as sometimes the volume gaussians will do a sequence of "2B 2R 2B 2R" indefinitely until such time the valid existence of either B2B 2R 2B or R2R 2B 2R occurs again and is identifiable as such?

 

2.b) The diagram of gaussians on Page 1 of this thread shows R B2B 2R. Should there be a 2B on the end of that sequence, but happens not to be shown in the diagram? And the R in the first position of that diagram isn't part of "B2B 2R 2B"?

 

2.c) Is there some minimum requirement, on a given chart magnitude, for a gaussian to be valid? Let's say we're looking at a 5min chart, and so we can draw a line w/out doing hocus pocus to come up with our conclusion? Is it two same color vol bars in a row, with the 2nd having to be either increasing or decreasing so that our arrow/line makes sense in that direction? Or is it 2 of 3 bars (in any order) which have to be the same color? Or 3 of 4 bars, or any other general pattern (even if it's just "usually" the case & not a hard rule)? This question is based on the assumption that the price chart isn't completely contrary to what we're trying to identify, since I know that goes hand-in-hand with volume gaussian recognition.

 

I do understand that the gaussian sequence pattern may exist on one magnitude/timeframe even when looking at the same chart. I'm moreso asking about identification when we're looking at the same level of granularity (more as a basis for differentiation/understanding).

 

Overall, I'm a little bit amazed so far at the idea that there is a never-ending flow of at least 3 price touches to a channel or traverse, no matter what? I understand trend lines, chart patterns and such, from past studies, but I never really focused on whether there was a repeatable pattern of that nature which was true 100% of the time as it seems this thread is claiming. If so, that's a huge bonus towards proper market behavior identification. If this is the case, then I assume sometimes the channel or traverse can and will be very very slight as far as the slope of it? Since, sometimes the market is sideways for days or weeks at a time and our traversing of that 'almost parallel' channel would basically amount to the price action being range-bound as it bounces back and forth within the channel? But, if we correctly identify and play the touches (as related to volume), then we can still have success in what others are calling a non-trending market. Right?

[ptunic 10]

Thanks a lot for the thoughtful replies, ptunic! I appreciate that. I'll ask some follow-up questions as below, for whevenever/if you get some time to share again. Did you take a break from this because you were having too hard a time defining (differentiation) things clearly enough to make it actionable?

 

Welcome!

 

The short answer is yes. While I made progress and my charts “I think” were getting more accurate, at a certain point I felt I would really have to dedicate even more time and energy to be successful. I had various life events going on (changing job, major relocation) and lost a lot of momentum and haven’t been learning trading since then. I work on mostly video game development and network programming projects for my main side projects these days. But, I enjoyed the learning experience of this method quite a bit, and you never know, may return to it one day.

 

1) FTT = failure to traverse; thanks for the clarification. So then, in a vanilla example we'd get a point 3 and then a FTT afterwards at some point. But you mentioned a point 5, and so on. Does that means the 1-2-3 has started over? Or is it a 1-2-3-4-5-6-n sequence when the trend continues? If so, I thought Spyder said that the fractals or sequence was always the same, without exception? I'm probably just not understanding exactly what was meant. Your clarifying this helped me (thanks) "b2b-2r-2b-FTT -> r2r-2b-2r-FTT etc"

 

Check out page 1, post #3 on this thread and you will see examples (I’ve taken it and edited the image and attached it here too). The “VE” (Volatility Expansion) can be considered Point 4 which has to be formed on increasing volume. Price will next fall in the non-dominant direction on decreasing volume. Then, volume will increase again in the dominant direction, and the trough of that volume is where your Point 5 is. These 2 new points have created a new, accelerated channel. Now either 2 things happen, you get a FTT which is a Point 1 of a new sequence in the opposite direction. Or.. price traverses the new channel, creating yet another VE (a Point 6). Seem familiar?

 

Another way to this is instead of Point 4 and Point 5, that the original Point 3 has now transformed itself into a brand new Point 1 of a new accelerated channel. Point 4 in this line of thinking is really a new Point 2, and Point 5 is really a new Point 3.

 

Whenever you create a new channel, one of the first questions you should strive to answer is what fractal is it? Is this new channel on the same fractal as I was just on? Or 1 size bigger or 1 size smaller? In the VE case, I think they are always the same fractal as the original (unaccelerated) channel.

 

 

2) For the Gaussians, and the sequence of either B2B 2R 2B or R2R 2B 2R, is the repeating pattern (or fractal) the simple fact that there will be a channel or traverse with a series of touches?

Not quite sure I understand but I’d would say the main thing is the volume patterns (Gaussians) much match.

 

up channel =

B2B = decreasing black volume followed by increasing black volume

2R = decreasing red volume

2B = increasing black volume

 

A note is decreasing volume can sometimes manifest itself as sideways movement as opposed to price movement in the opposite direction of the channel.

 

Keep in mind, not a hard and fast number but 90% of this system’s accuracy comes from the Volume pane, only 10% comes from information on the Price pane. Look at the very first sentence of this thread “Volume leads Price. Always. And without exception.” Notice he does not say Price leads Price, but Volume leads Price.

 

On a related note, let’s say you are looking at a chart and you look at just the Price pane (a big mistake!) and you see what “looks” like a crystal clear up channel with 3 points etc. Then you look at the Volume pane and you see the volume patterns do not match. That always means you have misidentified the channel. It could be just a subset of a different channel or a different fractal than you had in mind etc. But it is impossible to have an up channel without this volume pattern in this system. The peaks and troughs in the Volume pane are really what identify channels.

 

Once you get to a certain amount of experience in annotating charts, you’ll even quickly recognize this and self-identify this in your own charts. I have some old charts I’m sure years ago posted somewhere in this thread that I added a note onto a prospective channel saying something like “my Gaussians don’t match! Thus this channel is clearly wrong, but I can’t reconcile it with the surrounding channels”. This is actually pretty common to run into for many students.

 

 

2.a) Is there a way to see B2B 2R 2B 2R 2B 2R as something else, and be correct in seeing it? Or is it as simple as sometimes the volume gaussians will do a sequence of "2B 2R 2B 2R" indefinitely until such time the valid existence of either B2B 2R 2B or R2R 2B 2R occurs again and is identifiable as such?

I’m not quite sure what you mean here, but basically the main thing is you have 3 points (at least) in a sequence. It may end with a FTT or it may continue (sometimes for a whole day) if you continue to see the Gaussians showing decreasing non-dominant volume and increasing dominant volume. Eg in an up-channel, if the whole day (which happens on rare occasion) all the black volume bars are increasing and all the red bars are at lower volume than the previous bars, the trend hasn’t changed.

 

 

 

2.b) The diagram of gaussians on Page 1 of this thread shows R B2B 2R. Should there be a 2B on the end of that sequence, but happens not to be shown in the diagram?

Yes, it is just cut-off there, but if you extended that chart there would certainly have to be a 2B there.

 

And the R in the first position of that diagram isn't part of "B2B 2R 2B”?

Correct. That first R is the final dominant movement (2R) of the previous channel.

 

2.c) Is there some minimum requirement, on a given chart magnitude, for a gaussian to be valid? Let's say we're looking at a 5min chart, and so we can draw a line w/out doing hocus pocus to come up with our conclusion? Is it two same color vol bars in a row, with the 2nd having to be either increasing or decreasing so that our arrow/line makes sense in that direction? Or is it 2 of 3 bars (in any order) which have to be the same color? Or 3 of 4 bars, or any other general pattern (even if it's just "usually" the case & not a hard rule)? This question is based on the assumption that the price chart isn't completely contrary to what we're trying to identify, since I know that goes hand-in-hand with volume gaussian recognition.

There are many subtleties here, you cannot always identify the Gaussians exactly by just looking at a few bars. The general Gaussian logic holds and is the best starting place for interpreting charts, but you can’t break it down into just a small number of simple rules like “3 increasing red bars on the volume pane for sure mean xyz”. You have to identify what fractal you are on, which is not an easy task to do consistently, etc.

 

This is where some of the challenge to learning this method lies. .

 

 

I do understand that the gaussian sequence pattern may exist on one magnitude/timeframe even when looking at the same chart. I'm moreso asking about identification when we're looking at the same level of granularity (more as a basis for differentiation/understanding).

 

Fairly quickly in the process, you have to look at multiple timeframes (fractals) at once however. At the very least (and often times this is all you need to do), that is specifically 3 levels of fractals. This becomes important because understanding what is going on with the immediately larger and immediately smaller fractals at any particular point can give you dramatically clearer understanding of what is going on. I believe in many cases it is actually even a requirement of getting the correct result.

 

All that said, when first starting out, it is easier to as you suggest, it probably is easier to try with just a single level of granularity, and just don’t worry about “getting stuck” so to speak and move on to the next channel when one doesn’t make sense. Once you can add 3 fractals at once on a chart things will become clearer so that is the direction to strive towards.

 

Overall, I'm a little bit amazed so far at the idea that there is a never-ending flow of at least 3 price touches to a channel or traverse, no matter what?

Yes, it’s pretty cool When I randomly look at a stock chart for a few seconds now (even something as lame as Yahoo’s default quotes without the good charting software), my eyes can’t help but try to visualize channels, aligning them with Gaussians. It’s kind of like playing or composing music, it is like your brain just goes into a new mode and you see things that many other people won’t see.

 

I understand trend lines, chart patterns and such, from past studies, but I never really focused on whether there was a repeatable pattern of that nature which was true 100% of the time as it seems this thread is claiming. If so, that's a huge bonus towards proper market behavior identification.

Indeed it is. The accuracy the top traders of this system have is spectacular and far superior to anything else I’ve seen.

 

If this is the case, then I assume sometimes the channel or traverse can and will be very very slight as far as the slope of it? Since, sometimes the market is sideways for days or weeks at a time and our traversing of that 'almost parallel' channel would basically amount to the price action being range-bound as it bounces back and forth within the channel? But, if we correctly identify and play the touches (as related to volume), then we can still have success in what others are calling a non-trending market. Right?

Well, not sure what you mean here— some days you will see mostly very flat looking channels and other days channels that occur at very high or very low angles. You will see channels that are very wide and channels that are pretty tiny.

 

Along the same lines of this, the higher the volume, the easier it is to annotate (and thus trade) with this system as well as generally more profitable. I remember back in 2008 around November when there were market drops of 5% and more, some of the traders using this system had some of their best days ever. Some traders of this system don’t even trade except on days they know will have heavy volume (Federal Reserve decisions, unemployment #’s etc) as they will probably making 3-5 times more money those days than average. The other days can be spent at the beach

[Logic 10]

Is there any coding for any of Hershey's or Spyder's stuff? I'd like to test it using range bars. Thanks in advance.

[plantrader 10]

@ptunic: Thanks for the reply! I'm digesting it and making sure my next follow-up questions make sense. Or, as close to making sense as I can come at this point. lol! :crap:

 

Just one for now... are there any caveats such as the volume gaussian not having to fit any identifiable pattern during those timeframes where the price is within a lateral formation? The reason I ask is, I notice during flat price action the volume bars are often alternating colors back and forth. Or, they take turns with which color has higher volume.

 

The past 3 days of market action, I've had a harder time identifying recognizable volume gaussians, due to them being so mixed during less volatile price behavior. And in each case of identifying something, it seems there's almost an equal oppty to identify the opposite. So I thought maybe I was missing some general premise re: how to identify the sequence.

 

Oh, this would help me to know... what is the minimum # of volume bars needed for a gaussian line to be drawn (hiking up/down mountain)? Two bars? Or is it always several bars? Or can it actually be 1 bar to start & complete a gaussian up or down line all at the same time?

 

I'll save my other questions, since if I get a better grip on these questions then I should be able to simplify my upcoming questions a little bit better.

 

-plantrader

[ptunic 10]

Is there any coding for any of Hershey's or Spyder's stuff?

 

In terms of coding, not sure if you are referring to tools to assist a trader to make it easier to annotate charts and trade real time. Or if you mean algorithms to "back-test" the market where this method is coded into deterministic math.

 

For the first, yes, Spydertrader and some of the students worked together to create a special Trade Navigator suite. It does a lot of things for you including highlight the bars red or black correctly, mark lateral formations, and I think a few more things like Peak Volume and IBGS. I forget how to get that setup, I think the instructions were in an Elitetrader thread called something like Tools to Trade Price/Volume or something like that. It was the biggest thread in their software section so if you sort from views it was #1, but I haven't been on that site more than a couple times in many years so not sure if they deleted it or not. There are other software addons for this besides Trade Navigator, but that is the most definitive and complete tool and where most of the effort when into last time I was doing this.

 

In terms of the second, the system is deterministic and thus possible to codify. It often looks subjective only in the sense that most students have not mastered it, but for people close to Spydertrader's level they typically produce identical charts when they fully annotating them. At least that is my understanding of it. There are several efforts to codify the system and trade it by fully automated expert system but I'm not in the loop on these.

 

I'd like to test it using range bars.

I don't think it is compatible with range bars.

[ptunic 10]

Just one for now... are there any caveats such as the volume gaussian not having to fit any identifiable pattern during those timeframes where the price is within a lateral formation? The reason I ask is, I notice during flat price action the volume bars are often alternating colors back and forth. Or, they take turns with which color has higher volume.

Welcome! I'll just take a stab at this one for now.

 

The main concept is when you are looking at volume Gaussians, it is not 1 simple mathematical pattern so to speak. (There is a math pattern, but it would be pages long)

 

The main idea is this-- during say Black Volume Gaussians (increasing or decreasing), that does not mean each bar is black! You can, and frequently have, Red Bars as part of Black Gaussians.

 

Black Gaussians do not mean every bar as part of them is Black. They just mean the overall movement is up but you still need to have context to identify them correctly.

 

Similarly, during lateral formations, you can somewhat treat those very small volume bars as a single Non-Dominant leg of the sequence. I'm grossly oversimplifying it, but the idea is you can have even 50 bars occurring in a lateral formation each at very low volume, half red, half black. For the fractal you are trading, you will probably be treating those as a single leg of non-dominant volume. Laterals are both extremely important and one of my main weaknesses. There are different possibilities to exiting laterals and I have not consistently identified these cases. There are several traders here who as soon as they see 3 bars of a lateral, they know which way it will exit.

 

Somewhere in here there is a "Lateral Drill" which is one of the most important posts. That's probably would I would work next on if I pick this up again.

 

 

Correction: Black Gaussians don't even mean that price is going up, prices can even be flat in a Non-Dominant Black Gaussian. That relates to lateral formations. But, for Dominant Gaussians, you need at least 1 bar where price is moving in the dominant direction and volume increases.

 

Where the subtleties come into play what does "increasing volume mean" exactly? This is where it gets into the subtleties. There are cases where 1 bar has higher volume than the previous bar and yet still does not "count" as "increasing volume" in terms of forming a Dominant Leg of the sequence. It often does count (and beginners should start with that), but there are cases where it also needs to be higher than the volume 2 bars ago. In other cases, even that isn't sufficient and it needs to additionally be higher another previous peak in volume.

Edited December 10, 2013 by ptunic

[plantrader 10]

The main idea is this-- during say Black Volume Gaussians (increasing or decreasing), that does not mean each bar is black! You can, and frequently have, Red Bars as part of Black Gaussians.

 

Black Gaussians do not mean every bar as part of them is Black. They just mean the overall movement is up but you still need to have context to identify them correctly.

 

Understood. But what if, in the midst of what looks like a decreasing black gaussian (after an increasing red gaussian looks like it may have ended a downtrend / possible FTT), we get a red vol bar which is higher vol than the previous black bar? Does that foul up our identification? Would we be told later, by an expert, that we ignored that clue which invalidated our identification of our decreasing black gaussian?

 

Similarly, during lateral formations, you can somewhat treat those very small volume bars as a single Non-Dominant leg of the sequence. I'm grossly oversimplifying it, but the idea is you can have even 50 bars occurring in a lateral formation each at very low volume, half red, half black. For the fractal you are trading, you will probably be treating those as a single leg of non-dominant volume. Laterals are both extremely important and one of my main weaknesses. There are different possibilities to exiting laterals and I have not consistently identified these cases. There are several traders here who as soon as they see 3 bars of a lateral, they know which way it will exit.

 

Thanks. If, during a lateral formation after a dominant gaussian, there are bars of higher (in relative terms) volume in the non-dominant direction even though overall all the vol bars are fairly small during the lateral... does that somehow change the gaussian or our interpretation of it in relation to the existing gaussian which formed prior to the lateral? Or would the vol gaussian during the lateral still be considered part of the previous dominant gaussian? Could it be a clue that the lateral breakout will be in the non-dominant direction?

 

Or did I misunderstand, and the vol bars during a lateral should be identified as their own separate gaussian and we should attempt to draw the corresponding up/down mountain line as such? Even if doing that screws up the typical gaussian sequence of R BB R B or B RR B R? That's why I'm asking if the vol bars are part of the previous gaussian or if they're their own separate leg? Difficulty in interpreting that has caused me to get confused in my differentiation in real-time the past few days when there are quite a few laterals. Partially factoring-out the vol bars (re: gaussian identification) during laterals would have made it more clear to me at times.

 

Are the vol bars within a lateral less important and are just a continuation of the prior gaussian until such time the lateral is broken and vol increases? Then a new gaussian leg can be identified accordingly?

[plantrader 10]

One more quickie... would I be grasping at straws on this conclusion:

 

When aiming to differentiate & identify the gaussian legs, it's Ok to consider an increased volume bar of the "wrong color" as conforming to the overall gist of the that gaussian as long as the price action also conforms? For example, we're traversing upwards in price on a black dominant volume gaussian... then a bit of volume surges & we get a red volume bar BUT the price bar shows continuation of the pattern of higher lows. Therefore, the red volume bar can be considered as black as far as overall evaluation of that particular gaussian leg.

 

It might be a hint of what's to come, but with continued "higher lows" on the price bars... at this point it's only a hint/clue but not necessarily worthy of viewing the gaussian as invalid especially if the price action is fairly tight & non-volatile? Due to there being a fine line between black/red volume in such a circumstance. So as long as price conforms it's Ok in the near-term to see the gaussian as intact until further information comes in?

 

That, along with a similar viewpoint of flexibility during lateral price action, seems to have helped my perspective and frequency of being correct. That is, on a small sample size of course since I'm new to this. Probably an invalid sample from a statistical standpoint, as far as efficacy.

 

(p.s. The reason I ask this question: I utilized this conclusion/viewpoint on a Sim trade today & hit both targets and it played out very well)

[nive 10]

i'm used to doing everything by hand on esignal.i downloaded trade navigator and have problems getting the charts to look like i want. can anyone point me in the right direction on how to remove the overnight gap and coding the volume/price bars so they look right. i understand there is a premade library for that on there but i can't find it.

 

thank you very much.

[xioxxio 10]

i'm used to doing everything by hand on esignal.i downloaded trade navigator and have problems getting the charts to look like i want. can anyone point me in the right direction on how to remove the overnight gap and coding the volume/price bars so they look right. i understand there is a premade library for that on there but i can't find it.

 

thank you very much.

 

Somewhere in this thread there is an explanation. As far as I remember click the Telephone button (Update data from Genesis). In the Download special file section, type "PVTools" then Start

 

Otherwise call support, they will guide you.

[gltrader 10]

Wow, been awhile since I actually looked at these vids. Newbies can successfully refer to these vids to get the principles in place. In particular, find the cont, change video. It will really help you. This is an SCT thread, but the pace can be a bit quick for one unaccustomed to proper annotation.

 

Watch these videos, listen also very carefully to the questions asked in the background and how spyder addresses some of them.

 

Best Regards,

Oddi

 

Hello.

 

Can anybody help me get those videos? it would be a great help!

 

TIA

[gltrader 10]

Any comments are greatly apreciated

 

TIA

[Monkman 10]

Any comments are greatly apreciated

 

TIA

 

Added a little bit of clarity to your chart.

[plantrader 10]

Added a little bit of clarity to your chart.

 

Hi Monkman, in trying to continually understand this approach... could you clarify? You drew an up black volume gaussian (up the mountain line) which I've annotated in blue in my attachment. Why did you call that up black volume? I get it from the context that the price action was moving up at the time, but my understanding is that "volume drives price" and that volume gaussian patterns are what drive this. How did you see or identify that up black?

 

I know Spydertrader says this approach shows us that which exists, that we didn't know existed. But is it also sometimes seeing that which doesn't exist in order to FORCE the existence of the volume gaussian patterns? I see how you used that upward black to see in hindsight the RBBRB full sequence in hindsight, but in real-time how would you have ever seen that as black volume increase? My point is, sometimes it seems that the pattern doesn't exist on a particular level of granularity, but it might exist on a larger magnitude or vice versa. Is that true?

 

I'm probably missing a concept, but just asking the questions as it pertains to my current understanding. Set me straight, and thanks in advance!

[Monkman 10]

Hi Monkman, in trying to continually understand this approach... could you clarify? You drew an up black volume gaussian (up the mountain line) which I've annotated in blue in my attachment. Why did you call that up black volume? I get it from the context that the price action was moving up at the time, but my understanding is that "volume drives price" and that volume gaussian patterns are what drive this. How did you see or identify that up black?

 

I know Spydertrader says this approach shows us that which exists, that we didn't know existed. But is it also sometimes seeing that which doesn't exist in order to FORCE the existence of the volume gaussian patterns? I see how you used that upward black to see in hindsight the RBBRB full sequence in hindsight, but in real-time how would you have ever seen that as black volume increase? My point is, sometimes it seems that the pattern doesn't exist on a particular level of granularity, but it might exist on a larger magnitude or vice versa. Is that true?

 

I'm probably missing a concept, but just asking the questions as it pertains to my current understanding. Set me straight, and thanks in advance!

 

The annotations on that chart is what I would have done real time. There are certain aspects of the volume that I still do not understand well. Mainly under which type of channels do you need to see increasing volume to confirm the volume trend.

 

Lets assume I am wrong with my B2B annotations. Could the fractal of the volume annotations be off? Even if they were does it change how we represent the price movement? In further hindsight I can see describing the trend, see attached charts. There we visually have the B2B in terms of the length of volume bars. Now, the issue now is how do we represent the 2nd BBT. Is it a BBT? What about the third BBT, does the symmetric make it complex? Can the symmetric also be considered simple?

 

If you were to trade from one of these examples which picture would you pick to trade off of? Why?

[Monkman 10]

ES real time, b2b volume and price annotations

[Mandelbrot 10]

ES real time, b2b volume and price annotations

 

 

Next time try and get the image smaller. Why not show more of the chart?

[jbarnby 10]

The reality is your annotations indicate that you have jumped fractals. For date 01/02/14 the entire move to 1425 was a faster fractal, building the R2R of the fastest fractal we monitor. How does one know? Fractal integrity. One slower thing is built from three equal faster things. Pace acceleration also plays a role in fractal integrity, and it can do so on all fractal levels. (PA does not apply in this specific example)

 

Finally, the market achieves R2R or B2B in the exact same manner on all fractal levels, and it does so without exception. This has previously been covered in this thread.

[plantrader 10]

The reality is your annotations indicate that you have jumped fractals. For date 01/02/14 the entire move to 1425 was a faster fractal, building the R2R of the fastest fractal we monitor. How does one know? Fractal integrity. One slower thivolumeng is built from three equal faster things. Pace acceleration also plays a role in fractal integrity, and it can do so on all fractal levels. (PA does not apply in this specific example)

 

Finally, the market achieves R2R or B2B in the exact same manner on all fractal levels, and it does so without exception. This has previously been covered in this thread.

 

Hi jbarnby. Couple questions...I thought a lateral required 3 bars, but from your chart it looks like only two. Is two the rule, or three? Almost all the action for the day is contained in lateral the way you've drawn it. Also, how did you determine the "down black" volume line after the R2R? I don't see a decreasing black volume pattern. I do see the price action fits that pattern, though. That makes it look like you've used price to determine volume. Did you?

 

Am I being too stringent? From a technical standpoint, it seems sometimes the correct volume Gaussians do not exist in the expected pattern or order unless we concoct it to make it fit? Or that sometimes have to use "price only" for one leg the pattern because volume doesn't fit or isn't clear, which means the corresponding expected volume line is extrapolated in the context of the whole rb2b2r2b collection even though it may not exist for one leg? Such as the down black volume line you drew? I'm just trying to understand this better, any help would be appreciated.

Edited February 23, 2014 by plantrader
clarify my question

[Mandelbrot 10]

Hi jbarnby. Couple questions...I thought a lateral required 3 bars, but from your chart it looks like only two. Is two the rule, or three? Almost all the action for the day is contained in lateral the way you've drawn it. Also, how did you determine the "down black" volume line after the R2R? I don't see a decreasing black volume pattern. I do see the price action fits that pattern, though. That makes it look like you've used price to determine volume. Did you?

 

Am I being too stringent? From a technical standpoint, it seems sometimes the correct volume Gaussians do not exist in the expected pattern or order unless we concoct it to make it fit? Or that sometimes have to use "price only" for one leg the pattern because volume doesn't fit or isn't clear, which means the corresponding expected volume line is extrapolated in the context of the whole rb2b2r2b collection even though it may not exist for one leg? Such as the down black volume line you drew? I'm just trying to understand this better, any help would be appreciated.

 

 

I am not trying to be a smart ass, have you read this thread in entirety?

 

Laterals are 3 bars or more, other 2 bar formations are highlighted by the software.

 

Decreasing balck volume means black volume less than red.

 

A suggestion; read this thread and the futures journal on ET, only read spydertrader's posts no-one else. You will have a much better understanding of this process.