Trading The Wyckoff Way

[tupapa 10]

On the 23d of September, 1931 Wyckoff noticed an opportunity to go short after identifying climactic action on the 21st. Here is how he describes and justifies the short;

 

“On the 22nd, the volume drops off to about 2,000,000 shares; the close is slightly lower and the range has narrowed. The net result of these three sessions is to leave the market practically unchanged at the third day's close. Downward progress seems to have been checked and the small volume on the dip back from the high of the 21st, on Sept. 22nd, implies a lifting of selling pressure. After such a great decline within three weeks, this is an indication of more rally. This comes on the 23rd, and gives us an opportunity to sell short again while the market is still strong or when we see the rally is failing. Such an indication is given by the way it rallies on the 23rd. On this day, the average recovers to nearly 107, closing at 105½, but the volume falls off to under 3,000,000 shares and we therefore suspect that it is merely due to shorts who all tried to cover at once. Such a rally is too effervescent. It is not likely to last because it removes buying power which formerly existed, and leaves the market without support between the high point of the rally and the previous low.”

 

I am not the brightest of guys and it’s taking me some time to get my head around this, so here are my doubts;

 

1- He says that on the 23d volume falls but I, in fact see an increase in volume from the 22nd, and would consider this a bullish sign. The volume is lower than the climax, but then I wouldn’t expect it to be higher.

 

 

2- The range on the 23d is very wide, indicating little selling pressure and high buying pressure, such a rapid rally leaves the market with no support.

 

• Ok, I think I get this, so if the buying was of “good quality” buyers wouldn’t bid prices up right? They would rather take everything that’s on offer around the climax lows?
   
  
• What would the chart look like had the buying been of good quality?
  

 

I have attached an image of the chart up to the 23rd, for those of you interested on how the story unfolds, please refer to the first post on this thread or to section 7 of the original course, I am trying to understand how Wyckoff knows by the 23d, that the climactic action on the 21st will probably lead to a continuation rather than to a reversal.

Edited June 11, 2012 by tupapa

[DbPhoenix 42]

Very nice Db, thanks for the post, I hadn't considered what traders using different bar intervals would be looking at and how it would influence their decision making...

 

I'd never heard about information risk, price risk, and time risk, could you let me know where I can read more about this?

 

Thanks,

 

Tupapa.

 

Few people have. They think only of price risk, but it's largely the information risk that's generating their fear. Since they don't know the difference, they don't know what to do about it.

 

There used to be a guy here on this site that waited for so much confirmation after trend reversals that by the time he finally entered he was stopped out almost immediately because he was so late. He didn't like the fact that I pointed this out more than once. But now he's a bigtime guru charging big bucks to teach newcomers "how to trade". Yeah.

 

Anyway, the book is The Nature of Risk. Here's my review of it:

 

53 of 53 people found the following review helpful

Epiphany, anyone? January 14, 2000

By dbphoenix

Format:Paperback

The Nature of Risk is a seminal work for anyone who understands that self-knowledge is key for success in the financial markets, particularly at market extremes. Rather than babble about risk in general, Mamis takes this engine apart and examines its parts, among which are information risk and price risk. He explains that as one's tolerance for information risk increases (the need to know why the stock is doing whatever it's doing), one's price risk diminishes (one is better able to jump in and take advantage of whatever opportunities for picking up cheaper shares present themselves). On the other hand, if one has no tolerance for information risk and must know everything about a stock's movement, his price risk will be that much greater because the price will likely, by then, have risen to an over-extended level. Therefore, having identified these components of risk (time risk is another), one must then balance them out in order to approach the markets rationally and unemotionally.

 

An extremely important work, particularly for the investor who is plagued by doubt, confusion, and anxiety.

 

Db

[tupapa 10]

You're going to have to develop a feel for the turn. This is most easily seen with a very small interval. You may see volume increase or decrease as price reaches R. You will likely see the spread narrow dramatically, as in your chart. You'll detect a struggle to push through, followed by a rapid and decisive failure. If you place a sellstop below that last, narrow bar, then you will be swept into the short by the failure. If there is no failure, then your trade isn't triggered.

 

One last thing on this and I promise to move on from this trade. In the attached chart I have reduced the volume when price approaches the Resistance level.

 

If the volume wasn't climatic but price still reacts in the same way, would the sellstop below the last narrow bar be appropriate?

 

I am attaching the chart with the reduction in volume.

 

Many thanks.

[DbPhoenix 42]

One last thing on this and I promise to move on from this trade. In the attached chart I have reduced the volume when price approaches the Resistance level.

 

If the volume wasn't climatic but price still reacts in the same way, would the sellstop below the last narrow bar be appropriate?

 

I am attaching the chart with the reduction in volume.

 

Many thanks.

 

Information regarding trading activity is provided by the level of volume and by price action (or price movement). While the information provided by volume may be interesting and may help the trader reach a trading decision, it is itself largely irrelevant. What matters most is the price action.

 

Here, you're at R. Price chokes. That's really all you need to know to make your decision.

 

Keep in mind that if price fails, you're in the trade. If it doesn't, you're not. What's the risk?

 

Db

[tupapa 10]

I find this quite disconcerting... My reversal pattern is climatic action at Support/Resistance, followed by a re-test or Higher/Lower Low on lower volume.

 

 

However, from your post, I gather that as long as price is rejected at S/R the volume is irrelevant?

 

When, if anytime, should the whyckoff daytrader pay attention to volume?

[joshdance 10]

I find this quite disconcerting... My reversal pattern is climatic action at Support/Resistance, followed by a re-test or Higher/Lower Low on lower volume.

 

However, from your post, I gather that as long as price is rejected at S/R the volume is irrelevant?

 

That's a very valid pattern, but have you ever made or lost money based on volume increases or decreases? What matters is where the price is going. Volume can help and I use it extensively, but if you find yourself with your eyes more on the volume than the price, then you're focused in the wrong place. DB has well illustrated this point.

 

Regarding price vs. information risk: the earlier you get in, the less confirmation you have, but the better price you can get (risking less information, for a better price, is information risk); the later you get in, you have more confirmation, but the price is worse (risking a worse price, and more monetary risk on this single trade, while having more information to make your decision, is price risk).

[DbPhoenix 42]

I find this quite disconcerting... My reversal pattern is climatic action at Support/Resistance, followed by a re-test or Higher/Lower Low on lower volume.

 

 

However, from your post, I gather that as long as price is rejected at S/R the volume is irrelevant?

 

When, if anytime, should the whyckoff daytrader pay attention to volume?

 

Over the past half-dozen years or so, Wyckoff has become a racket. The complexity of the merchandising and promotion are such that one would not be surprised to learn that (fill in the massive American conglomerate of your choice) was behind it. Wyckoff knockoffs are nothing new. They go back decades. But with the nineties came the personal computer, the web, and their mutant offspring, the retail daytrader, a vast population of new victims ripe for plucking.

 

Given this situation, and the not uncommon desire on the part of the trader to put forth as little effort as possible, it was only a matter of time before a paint-by-number approach would develop and gain popularity: this volume bar means this and this price bar means that and put them together and you have a winning trade. Simple. Um hmm.

 

But in the real Wyckoff world, it doesn't work that way. Volume is not an indicator. Volume reflects transactions. As such, it is a piece of the puzzle, like the time of day or the day of the week or the general market environment. But there's nothing mysterious about it. And in and of itself, it does not prompt any particular action. What difference does it make, after all, what volume is if price is not doing anything that triggers a trade? And what if one's volume data were inaccurate and unreliable? What if volume weren't available at all? What if one were charting P&F?

 

"Climactic" volume may accompany climactic behavior in price or it may not. It may occur several days (or more) earlier. Or it may occur later. Or not at all. Or it may occur while price just shrugs its shoulders. In any case, it's just information about trader activity. Its weakness or near-absence can be as informative as its more dramatic moves if only one focuses on the information being provided and does not try to discern the extraordinary MEANING that at least one knockoff assigns to it.

 

In your case, having lower volume on a retest is nice, but what does it mean? Answer that and you'll know how much weight to give it, if any. As for when the daytrader -- Wyckoff or otherwise -- should pay attention to volume, given what's been done with it over the past dozen years with the color-coding and that silly experiment with the fat bars and all the efforts to make it into an indicator of one sort or another (all of which require software and/or special feeds and more money), I'm inclined to tell the beginning trader to just skip it, or at least to postpone its use until he has a thorough understanding of the hows, whys, whens, wheres, and whats of price movement. At the very least, this might prevent him from deciding not to take a perfectly good trade for no other reason than that the volume wasn't "right".

 

Db

[tupapa 10]

Thanks to both for your replies, I'd like to clarify that I do not focus on volume but on price. However I thought volume could be a valuable addition, and one that would increase the probability of success of a particular setup (in this case a reversal).

 

In your case, having lower volume on a retest is nice, but what does it mean? Answer that and you'll know how much weight to give it, if any

 

In my case, having lower volume on the retest means traders a less interested in this price level, and the probability of price bouncing, in search for a zone where traders feel comfortable increases.

 

One question for both of you;

 

If you see a test of support on average volume, and volume increases on the retest, would you still buy the reversal?

 

DB, I don't use coloured volume or any of the fat bars you mention, I just use plain voume bars showing market activity. I use them to find out the "relative interest of traders" in a particular price.

[joshdance 10]

If you see a test of support on average volume, and volume increases on the retest, would you still buy the reversal?

 

This depends on the context so much that it is impossible for me to answer yes or no. The first chart attached shows ES this morning. The dark yellow area to the left is the overnight session. You can see that the area highlighted had provided solid resistance for the whole session. Yet, I was long at the time, and when the market tested this area during the morning session I saw heavy volume, and the market immediately traded down pretty quickly. I exited my longs, and the quick and strong push to break through and subsequent immediate failure to do so caused me to rethink my position. But had the volume been very strong, and then broken through, then that would be another scenario.

 

The second chart shows a scenario that you described earlier. A retest of a high or low (high in this case) on much much lower volume. I was long at this time, and confess that I did scale the second of my three units when the market failed to break through the high convincingly. But no way I would short it. You can see what happened later if you want -- a nice squeeze up for another 6 or so points into the close, and I exited during this time to flatten. So, was there less "interest"? Yes, but there was more buying pressure than selling pressure, as evidenced by the fact that the market drove another 6 points into the close. There may have been less interest initially in buying above the high, but there was also less interest in selling...

 

So, despite what the volume is telling you, what is the price telling you? Is this a good short opportunity, in THIS context? Well, we have an early day test below yesterday's low, (and yesterday was a huge selling day) a rejection, and subsequent 15 point move almost straight up. Does this market want to be long? In my opinion, yes, which is why I was long. So, am I going to short the high on the retest, even though it's on lower volume? No way in hell--we have higher lows, one bull flag after another, into the market's favorite time to screw people who love to fade: 3:30 to 4:15pm. My basis was 09.50, I had scaled, and was risk free with a stop at 08.00 -- the market would need to trade down there to change my bias. The market must prove that its sentiment is changing, and low volume in this context does not offer any such proof to me. Price offered proof.

 

A lot of reading the market is due simply to looking at how it is behaving. Watching the bars/lines/candles/whatever as they move. This can be deceptive sometimes, but always tells a story different than a static chart. The quick rejection off the premarket highs as in the first chart was pretty clear evidence that I did not want to be long in the short term--it was the WAY it was rejected. However, the second chart, while showing an initial rejection of the highs, when observed in real time, showed no such urgency, or immediacy.

[BlueHorseshoe 10]

 

"Climactic" volume may accompany climactic behavior in price or it may not. It may occur several days (or more) earlier. Or it may occur later. Or not at all. Or it may occur while price just shrugs its shoulders. In any case, it's just information about trader activity. Its weakness or near-absence can be as informative as its more dramatic moves if only one focuses on the information being provided and does not try to discern the extraordinary MEANING that at least one knockoff assigns to it.

 

Db

 

I think the other thing that a lot of Wyckoff educators do is fail to distinguish between the way volume signifies in different markets. I would contend that volume in the indices, for instance, is very different to volume in currencies in terms of what it can tell us. This kind of blanket approach, promoted by many educators, of applying a strategy (or expensive, exclusive indicator) to any and every market is misleading, I feel.

 

BlueHorseshoe

[tupapa 10]

Some levels for today.

 

[Note: I moved your post here because you make no mention of what you're looking for or what you plan to do. If you'd like to add all that, I can move it back to the Foresight thread. Db]

Edited June 13, 2012 by DbPhoenix
Added note

[tupapa 10]

Thanks for the post Josh, I understand what you are trying to get across and I am aware of the importance of recent and historical context when making a trading decision.

 

I think my problem is I try to perceive price and volume in a very mechanical and objective way but there is a reason for this;

 

I am backtesting the reversal setup I mentioned (Climactic action followed by lower volume on the re-test) in order to estimate the probabilities of success and I need to do this in a purely mechanical way.

 

As I gain experience I guess I'll learn to trade more subjectively, thanks again for the reply.

[DbPhoenix 42]

Thanks for the post Josh, I understand what you are trying to get across and I am aware of the importance of recent and historical context when making a trading decision.

 

I think my problem is I try to perceive price and volume in a very mechanical and objective way but there is a reason for this;

 

I am backtesting the reversal setup I mentioned (Climactic action followed by lower volume on the re-test) in order to estimate the probabilities of success and I need to do this in a purely mechanical way.

 

As I gain experience I guess I'll learn to trade more subjectively, thanks again for the reply.

 

Keep in mind that when you're exploring possible "setups", the purpose of the backtesting is largely to find out if there's anything worth pursuing. The meat of this process comes in the forwardtesting, preferably with replay, so that you can see the bars form. It's so much easier to get a sense of traders' intentions when you can see the price move rather than review a series of static images. And, after all, dynamic price movement is what you'll be viewing when you trade; getting used to it is part of the function of forwardtesting.

 

Making it mechanical is always an option. Encouraging a beginner to exercise his judgement at the beginning is arguably unreasonable since he has so little to base his judgement on. Once you've viewed many charts, you'll be more comfortable with an increasingly subjective approach. But for now, it's perfectly fine to look for a particular pattern and trade only that pattern. If it works for you, you'll build your confidence, and that matters more than getting everything right from the getgo.

 

When you're backtesting, look for a pattern that repeats itself with reasonable consistency but also with enough frequency to make it worth waiting for. Then forwardtest it to see how much of the profitability was in the action and how much was in your head (hindsight often makes trades better than they were at the time, but that's how the gurus fleece the innocent). A pattern that is based on trader behavior won't change much because trader behavior doesn't change much.

 

The example you provided, for example, is relatively common: a long bar reaching up toward R, then price runs out of steam, sellers can't find any buyers, a short bar ensues, then it all rolls over. Look for that pattern and see how it works for you. It may not be entirely mechanical, but it may be close enough, at least close enough to formulate a few rules (including a rule that if the rollover is aborted and buyers make another run at R, the apparent weakness may not be real, and the appropriate response may be to back off).

 

Leaving mechanical trading for a lot of fuzzy "it depends go with the flow" babble -- however legitimate the babble eventually turns out to be -- is too big a step for just about anybody. Begin by focusing on what's in front of you, independent of what you want or expect, rather than forcing whatever you're observing into a preconceived set of expectations (like, for example, Fib, or Pivot Points, or moving averages). You will then have much less trouble finding a compromise between mechanical and subjective.

 

Db

[joshdance 10]

Thanks for the post Josh, I understand what you are trying to get across and I am aware of the importance of recent and historical context when making a trading decision.

 

Ok, but not only recent and historical context, but the context of this very day, of this very hour, is paramount and trumps all other context. I'm talking about the "right now" context, which unfortunately you cannot have by looking at old charts.

 

I think my problem is I try to perceive price and volume in a very mechanical and objective way but there is a reason for this;

 

I am backtesting the reversal setup I mentioned (Climactic action followed by lower volume on the re-test) in order to estimate the probabilities of success and I need to do this in a purely mechanical way.

 

As I gain experience I guess I'll learn to trade more subjectively, thanks again for the reply.

 

I just saw DB's last post, and I'll go a step or two further and be a bit more blunt: back testing does not work, and you are wasting your time in doing so.

 

There are too many variables that do not show up on your chart, such as news, day of the week, geopolitical events, and most importantly, you can't determine the market "mood" -- you can't simulate it, no matter how hard you try. Static charts are dead and lifeless, and we simply don't trade that way in real time.

 

Yes, the problem is you are trying to perceive price and volume in a very mechanical way, and they simply cannot be viewed that way, because they are not mechanical. Anyone who says they are, and who develops a mechanical way of dealing with them, will soon enough be changing their mechanical approach anyway, and they will repeat the cycle many times over, always looking for the mechanical answer which does not exist--so, you can go down that route if you like, but I decided not to.

 

There's nothing wrong with a visual examination of past data--but you should not find it "disconcerting" (in your words) that every retest of a high or low on lower volume does not turn into a reversal.

 

I know that sometimes I come across as blunt or possibly rude in my replies, so I hope you do not take offense in my perceived tone. If you could hear me say it, you would realize I'm only being straightforward. I would want others to be so blunt with me.

[wrbtrader 10]

Ok, but not only recent and historical context, but the context of this very day, of this very hour, is paramount and trumps all other context...

 

There are too many variables that do not show up on your chart, such as news, day of the week, geopolitical events, and most importantly, you can't determine the market "mood" -- you can't simulate it, no matter how hard you try. Static charts are dead and lifeless, and we simply don't trade that way in real time.

 

Strongly agree. Unfortunately, there's a growing group of coders and programmers that say a trader can easily code/program "market context". Yet, not one is able to produce such a code. :doh:

[DbPhoenix 42]

Ok, but not only recent and historical context, but the context of this very day, of this very hour, is paramount and trumps all other context. I'm talking about the "right now" context, which unfortunately you cannot have by looking at old charts.

 

Actually, with replay, you can. The patterns W finds in 1930-31 occur today as well, and will occur next week, month, year.

 

I just saw DB's last post, and I'll go a step or two further and be a bit more blunt: back testing does not work, and you are wasting your time in doing so.

 

I disagree. Backtesting, like simtrading, is useless if done improperly. Done properly, both backtesting (and forwardtesting) and simtrading are invaluable teaching and learning tools. The fact that few people do them properly is not an indictment of the tools.

 

There are too many variables that do not show up on your chart, such as news, day of the week, geopolitical events, and most importantly, you can't determine the market "mood" -- you can't simulate it, no matter how hard you try. Static charts are dead and lifeless, and we simply don't trade that way in real time.

 

I agree about static charts. As I've said many times, the market is a movie, not a slideshow. That's why I press the issue of replay. However, news and so forth are irrelevant. They color one's perception of the price movement, often to the degree that the trader assigns motive and meaning that are not in the movements but in his head. This can place him in a state of perpetual surprise.

 

Db

[joshdance 10]

I disagree. Backtesting, like simtrading, is useless if done improperly. Done properly, both backtesting (and forwardtesting) and simtrading are invaluable teaching and learning tools. The fact that few people do them properly is not an indictment of the tools.

 

To each his own, and I disagree but respect that others may find it useful. I never suggest that people do not backtest, as they may find the results more interesting than I have.

 

I agree about static charts. As I've said many times, the market is a movie, not a slideshow. That's why I press the issue of replay. However, news and so forth are irrelevant. They color one's perception of the price movement, often to the degree that the trader assigns motive and meaning that are not in the movements but in his head. This can place him in a state of perpetual surprise.

 

I think even replay does not simulate a live market. You cannot rewind or fast forward a live market, and the psychological control one perceives in a reply environment can carry a false sense of reality heading into a live market.

 

Regarding news, I don't mean to imply that if the news is good the market will go up. I mean that a surge in volume on a past chart must be checked with news for that day to ensure that the volume is not due to news. Heavy volume every first Friday of the month at 8:30am is always due to the same thing, thus heavy volume at that time is not significant. It should not be confused with volume resulting from normal trading activity, and one would have to check every past day's news events to ensure this if backtesting.

 

Happy trading today all!

[BlueHorseshoe 10]

I thought this thread seemed the perfect place to share something the market inflicted upon me recently that was so frustrating that I could only really laugh at it.

 

Rather than an individual trading decision, this relates to a strategy modification I recently implemented . . . I'm not a fan of hard stop losses (for my style of trading), and have written lots of posts on TL explaining why. Nevertheless, about six months ago I decided that if I could eliminate the possibility of occassional mammoth losses I would be willing to sacrifice some profit. After testing just about every stop-loss strategy that I could find or conceive, I settled on what seemed an acceptable solution, and began trading with a hard stop when I shorted the ES on 6th June

 

The result? Stopped out by one tick on 11th!!!

 

 

The second chart shows where my exit would have been without the hard stop, at the close of the same day, for a $437.5 per contract profit . . .

 

 

BlueHorseshoe

[DbPhoenix 42]

I think even replay does not simulate a live market. You cannot rewind or fast forward a live market, and the psychological control one perceives in a reply environment can carry a false sense of reality heading into a live market.

 

I agree with what you say after "and" (though not with what you say before), but that's not the point of back/forward testing as I incorporate it. Anyone who uses replay to back/forward test a mechanical system is likely to be disappointed, though I see people here and on other boards that have been doing so for up to ten years or more and still can't trade, much less turn a profit, so their capacity for disappointment may be limitless.

 

Price movement is fueled by fear and greed (and, arguably, hope, though hope is a form of fear). These emotions are manifested in trader behavior. The purpose of all this review and study in a Wyckoff context is to develop a sensitivity to this which will then enable the trader to anticipate movements. But if a trader thought it was all mumbo-jumbo, he wouldn't be studying W in the first place, none of which is to suggest that W is the only or even the best approach but only to explain why these means of learning the approach are investigated here.

 

Db

[BlueHorseshoe 10]

Strongly agree. Unfortunately, there's a growing group of coders and programmers that say a trader can easily code/program "market context". Yet, not one is able to produce such a code. :doh:

 

No?

 

You've not heard of, say, RenTech then?

 

Finding mechanical approaches that work is certainly not easy, and such approaches tend to work best when they contain some (mechanically) adaptive element, but it is by no means impossible to do.

 

This is not to say that the human brain cannot learn to do this better, but to dismiss mechanical trading in the way that you and Joshdance are suggesting is . . . well, dismissive.

 

BlueHorseshoe

[joshdance 10]

The result? Stopped out by one tick on 11th!!!

 

It looks like your strategy only trades RTH? The Sunday gap up was huge on the day you were stopped out and good at least that you did not take a stop 10 points higher where it gapped to!

[DbPhoenix 42]

I thought this thread seemed the perfect place to share something the market inflicted upon me recently that was so frustrating that I could only really laugh at it.

 

Rather than an individual trading decision, this relates to a strategy modification I recently implemented . . . I'm not a fan of hard stop losses (for my style of trading), and have written lots of posts on TL explaining why. Nevertheless, about six months ago I decided that if I could eliminate the possibility of occassional mammoth losses I would be willing to sacrifice some profit. After testing just about every stop-loss strategy that I could find or conceive, I settled on what seemed an acceptable solution, and began trading with a hard stop when I shorted the ES on 6th June

 

The result? Stopped out by one tick on 11th!!!

 

The second chart shows where my exit would have been without the hard stop, at the close of the same day, for a $437.5 per contract profit . . .

 

BlueHorseshoe

 

This may, however, be less an indictment of hard stops than an example of inappropriate entry. This is, after all, the Wyckoff Forum

 

Db

[BlueHorseshoe 10]

This may, however, be less an indictment of hard stops than an example of inappropriate entry. This is, after all, the Wyckoff Forum

 

Db

 

I know it is. So why not pull up a daily chart of the ES and show me how a Wyckoff-ian study of volume could have improved entry (or informed a better hard stop placement)?

 

And for the record, it wasn't intended as an indictment of hard stops - just a 'couldawouldashoulda' example of market perversity . . .

 

BlueHorseshoe

[BlueHorseshoe 10]

It looks like your strategy only trades RTH? The Sunday gap up was huge on the day you were stopped out and good at least that you did not take a stop 10 points higher where it gapped to!

 

Erm, actually I did!

 

I was stopped out within ten minutes of the gap opening on Sunday night, one tick from the 1342 high before the ES sold of into the trading day of 11th. Without the hard stop, I only entered or exited positions on the close, but now a hard stop is sitting there whenever the market is open.

 

BlueHorseshoe

[DbPhoenix 42]

I know it is. So why not pull up a daily chart of the ES and show me how a Wyckoff-ian study of volume could have improved entry (or informed a better hard stop placement)?

 

No point, as it would be only one more example of a hindsight trade with the accompanying hindsight insight one finds on DVDs and in webinars. And you can guess how much hindsight insight I've seen over 15 years on the Web.

 

It's not my purpose to persuade anyone of the superiority or even the efficacy of the Wyckoff approach but rather to help those who are genuinely interested in it to understand it and implement it. There are many paths to profitability, and this is only one of them. Lots of room on the bus for those who prefer some other route.

 

Db