You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Do Trendlines work?
- Thread starter duard
- Start date
Lamont C,
"However, this has much less to do with trendlines than it does with simple support and resistance.
When price approaches the "trendline", look to the left."
Indeed you are correct and if I might amplify my previous blather ever so slightly, please allow me to say that when testing one's theory, one best be sure that the "observable" one has selected is a meaningful observable. Without a meaningful observable you may not be testing what you think you are testing.
lj
"However, this has much less to do with trendlines than it does with simple support and resistance.
When price approaches the "trendline", look to the left."
Indeed you are correct and if I might amplify my previous blather ever so slightly, please allow me to say that when testing one's theory, one best be sure that the "observable" one has selected is a meaningful observable. Without a meaningful observable you may not be testing what you think you are testing.
lj
Quote from ljyoung:
Do you have any instances where in spite of your expectation that price would bounce off the trendline, it in fact didn't?
One need look no further than volume to locate the answers you seek.
In an uptrend, dominant volume (black) pushes Price higher. In other words, increasing volume (across the trend) leads to increasing price. The same phenomenon unfolds in a down trend where increasing red volume pushes Price lower. In other words, the dominant volume (red) when increasing (across the trend) leads to decreasing price. Retraces (decreasing volume and decreasing Price within a up trend [or decreasing volume and increasing Price within a down trend]) develop as price and volume move back to the right trend line. At the right trendline, increasing dominant Volume (in either up or down trend) leads to a return to the trend. Whereas, increasing non-dominate Volume (in the up or down trend) leads to a Price break out of the trend (what most term - a reversal).
As a result, a trader need not 'predict' the outcome of these events. By calculating Volume on a pro-rata basis (easier now that Medved's Quotetracker recently added an automated tool to provide the calculations), a trader simply needs to monitor the market in an effort to determine if a mode shift (continuation or change) is taking place. Once known, the choice becomes clear - hold, reverse or exit.
Good trading to you all.
- Spydertrader
Spydertrader,
Being well familiar with the Wyckoffian concepts associated with volume and price, I must say that in my experience there are several factors which come into play "at the rail", if you will. I do not use trend lines in my trading protocols but I am not saying that they cannot be useful to some traders.
I will preface my remarks with two general statements:
(1) If a small retail trader believes that there is something that he or she knows that the people who move the markets don't know, then in the highest of likelihoods they are deceiving themselves. That is not to say there is some massive conspiracy specifically directed at the hapless retail trader but rather that it is a reality of the market that big players move the markets. It is my belief that by observing what they are doing and understanding it, one has the best chance of surviving. As soon as you start trying to predict too far in advance what they are going to do, you will be devoured.
(2) Unless you are paying for and getting Tier 1 data from a reputable data source, there is a very high probability that what you are looking at in your chart is visual gibberish.
Moving along then, IMO the first matter of concern is the time frame in which one trades and the type of chart that one uses to trade with. I use both time and tic based charts contemporaneously. If the players aren't trading (the tic chart is not active) time is still active. As well you know when trying to decipher volume signals, the shorter the time frame (or tic frame) the more difficult it is. Next one needs to consider the average daily volume since highly liquid stocks are less subject to manipulation than thinly traded ones (an extreme example of course being the OTC pump and dump routine).
The color of the volume bars is not something I pay much attention to for several reasons. For time-based trading, most commonly the color is determined by the close of the bar in relation to the prior bar which of necessity means absolutely nothing with respect to whether there has been a preponderance of buying or selling within the time frame of the bar you are looking at. If you accept this premise then a possible workaround is to look at the tic chart but again there is the problem of deciding whether the trade which occurred between the best bid and best ask was a buy or a sell. A nice way to see this is to look at "detailed" a VbP profile, which QuoteTracker provides, and you will immediately see that the predominant color is "indeterminate". There is the further problem of how short do you make your tic interval. In the extreme lower limit of 1 tic/bar you are of course looking at a "visual" T&S readout. I know there are folks who make their living off T&S but my guess is that your methodology might prove to be a little hairy on application to that short a time frame.
So returning then to your statement of how volume can alleviate the confusion "at the rail", I would say the following. If we presume that you are trading an entity which has sufficient liquidity (and I am aware of Mr Hershey's thoughts about that aspect of things) then one is still left with the necessity of determining the best time/tic frame in which to trade in order that your volume signals make some sense. If one determines that for such and such an entity, a 30 minute time bar allows one to have a reasonable probability of getting things right, then cool. There is the further advantage that if your analysis turns out to occasionally be incorrect, you will know rather quickly whether you need to "hold, reverse or exit."
lj
Being well familiar with the Wyckoffian concepts associated with volume and price, I must say that in my experience there are several factors which come into play "at the rail", if you will. I do not use trend lines in my trading protocols but I am not saying that they cannot be useful to some traders.
I will preface my remarks with two general statements:
(1) If a small retail trader believes that there is something that he or she knows that the people who move the markets don't know, then in the highest of likelihoods they are deceiving themselves. That is not to say there is some massive conspiracy specifically directed at the hapless retail trader but rather that it is a reality of the market that big players move the markets. It is my belief that by observing what they are doing and understanding it, one has the best chance of surviving. As soon as you start trying to predict too far in advance what they are going to do, you will be devoured.
(2) Unless you are paying for and getting Tier 1 data from a reputable data source, there is a very high probability that what you are looking at in your chart is visual gibberish.
Moving along then, IMO the first matter of concern is the time frame in which one trades and the type of chart that one uses to trade with. I use both time and tic based charts contemporaneously. If the players aren't trading (the tic chart is not active) time is still active. As well you know when trying to decipher volume signals, the shorter the time frame (or tic frame) the more difficult it is. Next one needs to consider the average daily volume since highly liquid stocks are less subject to manipulation than thinly traded ones (an extreme example of course being the OTC pump and dump routine).
The color of the volume bars is not something I pay much attention to for several reasons. For time-based trading, most commonly the color is determined by the close of the bar in relation to the prior bar which of necessity means absolutely nothing with respect to whether there has been a preponderance of buying or selling within the time frame of the bar you are looking at. If you accept this premise then a possible workaround is to look at the tic chart but again there is the problem of deciding whether the trade which occurred between the best bid and best ask was a buy or a sell. A nice way to see this is to look at "detailed" a VbP profile, which QuoteTracker provides, and you will immediately see that the predominant color is "indeterminate". There is the further problem of how short do you make your tic interval. In the extreme lower limit of 1 tic/bar you are of course looking at a "visual" T&S readout. I know there are folks who make their living off T&S but my guess is that your methodology might prove to be a little hairy on application to that short a time frame.
So returning then to your statement of how volume can alleviate the confusion "at the rail", I would say the following. If we presume that you are trading an entity which has sufficient liquidity (and I am aware of Mr Hershey's thoughts about that aspect of things) then one is still left with the necessity of determining the best time/tic frame in which to trade in order that your volume signals make some sense. If one determines that for such and such an entity, a 30 minute time bar allows one to have a reasonable probability of getting things right, then cool. There is the further advantage that if your analysis turns out to occasionally be incorrect, you will know rather quickly whether you need to "hold, reverse or exit."
lj
Quote from ljyoung:
then one is still left with the necessity of determining the best time/tic frame in which to trade in order that your volume signals make some sense.
Actually, no. One need not determine which time frame to trade in order to make use of the PV relationship. The system I described above operates within a binary (not gradient) framework. In short, the process works in any market on any time frame - provided sufficient liquidity (and volatility) exist. From yearly charts to tic level charts, the process works the same exact way - only faster with each drop in time frame.
With respect to your 'prefaced' points above, these points have no influence on the PV relationship. Academic debates on the accuracy of data or the validity of information (or other 'edges') proves nothing more than an exercise in futility. However, one wishes to slice it, the PV Relationship effects the rate of price change at the rail, on the rail, next to the rail, above the rail or below the rail.
It works, not because someone says it works, but because this is the way all markets operate.
Good Trading to you all.
- Spydertrader
It was by and large Wyckoff who first got it right when he correctly interpreted the relation between price and volume and since his time there has been precious little to add to what he said about the matter except for "sequels"
It is not fatuous to state that if one's data is meaningless (visual gibberish) that there are no consequences with respect to determining the validity of a particular trading methodology. That is of course unless the outcomes are being determined by factors other than the variables being looked at, which in this case would be price, volume and time.
I'm ignorant as to what you contextually mean by binary vs gradient but would take you to mean that since the markets are fractal, then what happens on one time frame happens on all time frames. I could not agree with Mandelbrot more on that point. The way I constructed my view was to suggest that as the time or tic frame became shorter it became more difficult to accurately discern what meaningful relation existed between price and volume and equally important to make money while you are doing it. As I further said using your methodology on a 1 tick/bar chart of a highly liquid and volatile entity (SPY perhaps?) would prove to be rather hairy - not impossible, but hairy. Which is to say preferential order flow positioning and lots of computing power would be definite assets.
Thus it makes no sense to me to say that at very short time or tic frames that "... debates on the accuracy of data or the validity of information (or other 'edges') prove nothing more than an exercise in futility." given the exquisitely tight relationship between price and volume you have espoused in your post.
By extrapolation of your viewpoint then, the "colors" of the volume bars should be equally unimportant but clearly in your analysis they are not. For different reasons, outlined in the previous post, I do not believe the colors are significantly meaningful.
I am being a trifle hyperbolic in the immediately preceding paragraph but it is simply to make the point that at some point in your analysis you make a decision that the data you are looking at is indeed the data which is reflective of what is happening in the market and my question is when exactly is that?
I as yet am not knowledgeable enough to say that I know how all markets work but what I can say is that they are frequently not what they are portrayed to be (kind of like free trade with government subsidies). The role of trader ignorance, naivete and lack of understanding of what is really going on possibly contributes to the finding that some 90-95% of short term traders go bust. In so far as those folks were concerned the way the market worked for them was to relieve them of their discretionary funds with facility.
lj
It is not fatuous to state that if one's data is meaningless (visual gibberish) that there are no consequences with respect to determining the validity of a particular trading methodology. That is of course unless the outcomes are being determined by factors other than the variables being looked at, which in this case would be price, volume and time.
I'm ignorant as to what you contextually mean by binary vs gradient but would take you to mean that since the markets are fractal, then what happens on one time frame happens on all time frames. I could not agree with Mandelbrot more on that point. The way I constructed my view was to suggest that as the time or tic frame became shorter it became more difficult to accurately discern what meaningful relation existed between price and volume and equally important to make money while you are doing it. As I further said using your methodology on a 1 tick/bar chart of a highly liquid and volatile entity (SPY perhaps?) would prove to be rather hairy - not impossible, but hairy. Which is to say preferential order flow positioning and lots of computing power would be definite assets.
Thus it makes no sense to me to say that at very short time or tic frames that "... debates on the accuracy of data or the validity of information (or other 'edges') prove nothing more than an exercise in futility." given the exquisitely tight relationship between price and volume you have espoused in your post.
By extrapolation of your viewpoint then, the "colors" of the volume bars should be equally unimportant but clearly in your analysis they are not. For different reasons, outlined in the previous post, I do not believe the colors are significantly meaningful.
I am being a trifle hyperbolic in the immediately preceding paragraph but it is simply to make the point that at some point in your analysis you make a decision that the data you are looking at is indeed the data which is reflective of what is happening in the market and my question is when exactly is that?
I as yet am not knowledgeable enough to say that I know how all markets work but what I can say is that they are frequently not what they are portrayed to be (kind of like free trade with government subsidies). The role of trader ignorance, naivete and lack of understanding of what is really going on possibly contributes to the finding that some 90-95% of short term traders go bust. In so far as those folks were concerned the way the market worked for them was to relieve them of their discretionary funds with facility.
lj
I enjoy reading your posts and getting an understanding of where you ar coming from.Quote from ljyoung:
It was by and large Wyckoff who first got it right when he correctly interpreted the relation between price and volume and since his time there has been precious little to add to what he said about the matter except for "sequels"
It is not fatuous to state that if one's data is meaningless (visual gibberish) that there are no consequences with respect to determining the validity of a particular trading methodology. That is of course unless the outcomes are being determined by factors other than the variables being looked at, which in this case would be price, volume and time.
I'm ignorant as to what you contextually mean by binary vs gradient but would take you to mean that since the markets are fractal, then what happens on one time frame happens on all time frames. I could not agree with Mandelbrot more on that point. The way I constructed my view was to suggest that as the time or tic frame became shorter it became more difficult to accurately discern what meaningful relation existed between price and volume and equally important to make money while you are doing it. As I further said using your methodology on a 1 tick/bar chart of a highly liquid and volatile entity (SPY perhaps?) would prove to be rather hairy - not impossible, but hairy. Which is to say preferential order flow positioning and lots of computing power would be definite assets.
Thus it makes no sense to me to say that at very short time or tic frames that "... debates on the accuracy of data or the validity of information (or other 'edges') prove nothing more than an exercise in futility." given the exquisitely tight relationship between price and volume you have espoused in your post.
By extrapolation of your viewpoint then, the "colors" of the volume bars should be equally unimportant but clearly in your analysis they are not. For different reasons, outlined in the previous post, I do not believe the colors are significantly meaningful.
I am being a trifle hyperbolic in the immediately preceding paragraph but it is simply to make the point that at some point in your analysis you make a decision that the data you are looking at is indeed the data which is reflective of what is happening in the market and my question is when exactly is that?
I as yet am not knowledgeable enough to say that I know how all markets work but what I can say is that they are frequently not what they are portrayed to be (kind of like free trade with government subsidies). The role of trader ignorance, naivete and lack of understanding of what is really going on possibly contributes to the finding that some 90-95% of short term traders go bust. In so far as those folks were concerned the way the market worked for them was to relieve them of their discretionary funds with facility.
lj
It is really fun to continue to examine more and more closely what is going on.
I believe that you are recognizing by now that we are front running the markets.
the beautiy of understanding that there are several channels within channels and the is a direct correspondence between P an V for each of these channels as pairings keep everything very clean and precise. The set is a meaningful observation.
When the concurrent tick charts are added as OTR (One Tick Range charts (and their dwell volumes as well, there is no shift in reasoning it turns out.
This consistent and continuous phenomena you are beginning to observe is where SCT's name came from.
The sequences of the binary (and vector as well) perfomance clearly gives the user an advantage over those using powerful staticitcal analysis. What is it like to first begin to observe valuations where direction and magnitude (combined) values only have one alternative value.
Seeing a present value and knowing there is only one alternative for each of many components of a data set is a place to savor and cherish. All the names of people you mentioned have not had the experience that you are now having.
All the lines you see have definition in non probabalistic maths.
The degrees of fredom being used to montior are all binary vectors.
You may reflect and appreciate this place of operating that deals in only NOW and has no probabalistic factors.
There are no tests of significance needed since it is a priori.
you are so very correct and precise in requiring onlt meaningful observations.
there is only one kind: the binary vector, the only non-probabalistic value with zero uncertainty available continuously in, for this case the market.
Finite data sets. Finite and coresponding conclusion sets. Five decision alternatives that are permenantly associated with each of the elements of the finite conclusion set elements.
The market dictates a singular routine (and it is NOT OODA of Steenbarger); the routine is monitor for complete data sets, analyze (apply the predetermined conclusion element from a fintie set); decisde by using ut decision associated with the conclusion (predetermined as well).
by now you see a matrix appearing and it is finite. you will come to notice as you peruse it, that the market migrates as one consequence. That consequence is that, successively, all other pathways are closed off, and only one remains to be followed.
Is it difficult to sit and watch successive pathways closed off, mathematically. I think not.
Continuous data provide precision. Precision leads to effectiv8ieness and efficiency.
At some point the conventional paradigm in its probabalistic application becomes crude and insensitive.
Looking at the OTR and its collateral buddies (five in number) spreads the passage of time out so widely that it almost seems to stand still when decisions are made by taking timely action.
I really enjoy reading your stuff. You are very close to beginning to see how markets operate and best of all you can still consider what is possible. Stopping the consideration of what is probable is a really terrific experience.
Profitable traders exist in all markets - each using a variety of methods. Some find value in indicators, while others choose to monitor Price and Volume (or even simply Price alone). Whether employing an edge based strategy, some sort of arbitrage or simply monitoring the market in an effort to differentiate between the modes of continuation or change, each profitable trader has spent time internalizing his (or her) methodology. Can we make the same observation with respect to the 90-95 percent of individuals who no longer have the financial means to trade? Doubtful. People fail at many endeavors in life for a wide variety of reasons. We see no difference then between other industries and the trading industry. Some people fail, others don't.
I make no claims supporting or decrying the work of Wyckoff and Mandelbrot. While I am confident they are fine upstanding individuals, I have no reason to read or otherwise cite their work. How others view the market has no influence upon how I trade. I learned to trade, not from books, but by studying charts. I followed the charts because that was what successful traders (at the time I first began my Journey) advised me to do. They told me, "all you need is in the charts" and it turns out, they were right.
With respect to binary vs gradient, think of a light switch compared to a dimmer switch. One turns the light on and off, the other allows for various amounts of illumination. With respect to the binary nature of the PV relationship, it means it works or it doesn't. If it works, then the trader knows he (or she) has the correct trend. When it doesn't work, the trader knows he (or she) does not have the correct trend. Most often, this occurs when the trader has failed to 'see' the correct trend in a timely fashion. Once the trader has corrected their trend in such an environment, the PV Relationship again shows how price and Volume move the markets. Binary - it is or it isn't.
In an effort to avoid any confusion, with respect to the Price / Volume Relationship, it works on:
Any market, on Any time frame - provided sufficient liquidity (and volatility) exist."
Believe it. Don't believe it. Or, Debate it amongst your colleagues until the wee hours of the morning. It's your choice. As it turns out, all the answers anyone really needs reside "in the charts." I leave you now to continue the debate, and as always .....
Good trading to you.
- Spydertrader
I make no claims supporting or decrying the work of Wyckoff and Mandelbrot. While I am confident they are fine upstanding individuals, I have no reason to read or otherwise cite their work. How others view the market has no influence upon how I trade. I learned to trade, not from books, but by studying charts. I followed the charts because that was what successful traders (at the time I first began my Journey) advised me to do. They told me, "all you need is in the charts" and it turns out, they were right.
With respect to binary vs gradient, think of a light switch compared to a dimmer switch. One turns the light on and off, the other allows for various amounts of illumination. With respect to the binary nature of the PV relationship, it means it works or it doesn't. If it works, then the trader knows he (or she) has the correct trend. When it doesn't work, the trader knows he (or she) does not have the correct trend. Most often, this occurs when the trader has failed to 'see' the correct trend in a timely fashion. Once the trader has corrected their trend in such an environment, the PV Relationship again shows how price and Volume move the markets. Binary - it is or it isn't.
In an effort to avoid any confusion, with respect to the Price / Volume Relationship, it works on:
Any market, on Any time frame - provided sufficient liquidity (and volatility) exist."
Believe it. Don't believe it. Or, Debate it amongst your colleagues until the wee hours of the morning. It's your choice. As it turns out, all the answers anyone really needs reside "in the charts." I leave you now to continue the debate, and as always .....
Good trading to you.
- Spydertrader