Richard Wyckoff

In this post, I introduce Richard Wyckoff. Here’s how Wikipedia describes him (http://en.wikipedia.org/wiki/Richard_Wyckoff)

“Richard Demille Wyckoff (born November 2, 1873; died March 19, 1934) was a stock market authority, founder and onetime editor of the Magazine of Wall Street (founding it in 1907), and editor of Stock Market Technique……..”

But the write-up does little justice to a man who established a school of thought. Richard Wyckoff believed that understanding the context and principles of market movement was the path to success. His approach struck a responsive chord within me, fanning a spark that had been lit by Market Profile Theory. It’s strange how things work out – I came across Steidlmayer in 1980 and only later to Wyckoff’s (whose hey day was in the 1920’s). Yet the two works fit together like a seemless whole.

Many modern traders are unaware that Wyckoff”s approach was diametrically opposite to Richard Schabacker’s, the uncle of Robert Edwards (Edwards and Magee fame). Schabacker believed in the classification of patterns – the ‘why’ was less important than the ‘form’. I see Schabacker’s approach mirrored in many modern works. (For a write-up on Schabacker, see http://www.marketmasters.com.au/82.0.html)

Wyckoff opened my eyes to the relationship of:

  • Direction – which way is the market seeking to go?
  • Volume and range – how good a job is the market doing in moving in its desired direction.

Steidlmayer was later to call these factors ‘trade facilitation’.

Wyckoff’s idea behind ‘trade facilitation’ was simple: the effect (range) ought to mirror the cause (volume). If it failed to do this, the market was telling us that a new game was afoot. So, if we had above normal volume and below normal range, this was a warning of a possible change of direction.

While Wyckoff’s principles are easy enough to state, applying them (at least till now) was another matter. There were at least two troublesome questions:

  1. How do we determine the market’s attempted direction?
  2. How do we determine whether volume is net selling or buying for any given period?

Traditionally:

The answer to (1) is to relationship between the close of today and that of yesterday: e.g. a plus close was interpreted as an attempt by the market to move higher. But what happens when you have a day with higher highs and higher lows and a down close? I always found it diffcult in these circumstances to say that the market was looking to head South.

The answer to (2) was to treat a whole period’s volume (e.g. the day’s volume, the 30 mins volume etc) by referencing (1). If we had a down close, we’d treat the whole period’s volume as selling volume. You don’t need me to point out the logical flaws to this.

Once we had intra-day volume another problem arose. The issue is well-illustrated by the volume during an ES trading session. The greatest volume tends to take place in the first two hours. The volume then tapers off till the last two hours of trading when volume again increases – usually the volume for these last two hours is less than the first two hours. Figure 1 shows what I mean. The yellow rectangle shows one day’s trading. So, given this tendency, merely comparing one period’s volume with the previous one, was inadequate – we aren’t comparing apples with apples.

blog-12-20-2007-uv.jpg

FIGURE 1 ES 30 Mins Bars

Those were the problems. Tomorrow I’ll present the solutions.

Whither S&P?

I take the view that the S&P (ES H8) is on the cusp.

In “The US$ and the Stock Market”, I outlined my view of the fundamental context for the US stock market.

Technically: 

The 12-month swing shows that a close below 1455 would signal an Upthrust Change in Trend Sell Signal. See Figure 1.

 12m.jpg

FIGURE 1 12-M S&P Cash

The 13-week swing and Ray Wave show that the market is either in the process of forming a wave:4 Irregular.

Or it shows that the market has already triggered an Upthrust CIT pattern, and we need only see a Whole Point Count (WPC) to confirm (see Nature of Trends).  I have taken the former view and I am looking for one more high to complete the 5-wave structure. In this interpretation, I am influenced by the fundamental context.

Should the market break below 1370 (basis cash) and form a 13-w WPC, then I’d change my view.

Since the longer-term charts favour a continuation of the uptrend, I have been playing the S&Ps from the long side. On Nov 27th I went long and exited 1/3 positions on  Dec 10th. I brought my stops on 1/3 of the position under 1427.6 and 1/3 to breakeven. I based the stops on Market Profile theory:

  • The 1/3 exit was at the top of the Value Area (Zone 1 high).
  • The second 1/3 stop is under the Value Area(Zone 1 low);
  • The last 1/3’s stop at breakeven is to protect a position against loss where the environment for the long is deteriorating. Figure 3 shows the profile.

 18d-mkt-profile.jpg

FIGURE 3 18-d & Market Profile

This is where the market stands for me. I am not keen to add to my positions given my view that this is the last leg up and that there are doubts about the continuation of the uptrend.

But if I did want to take a position what would I look for?

  • The market is in a Sideways Market Zone (78.6% retracement) Figure 4

 18d-sw-zone.jpg

FIGURE 4 Sideways Zones

  • There was a cycle low on December 18th
  • Sentiment indicators (COT) is bullish (see below)
  • There are three statistical edges suggesting the low is in (see below for an example)
  • We had a Market Profile Test Day (read possible change of direction day). To confirm the test day, we need to see strong buying today.

Sentiment indicators and statistical data all suggest a rally from here. I use Floyd Upperman’s data for COT (https://www.upperman.com).

An example of statistical setup that is present answers this question: “How often would I be profitable if I bought on the close Monday (Dec 17) of option expiration week if the on that day the S&P hit at least a 2-week low and I held the position for one week”?

The answer is since 1950 we have had 6 winners from 7 trades. The average drawdown is -0.6% compared to an average gain of +2.2%. 

Statistical setups I gather from various sources: the one above from www.sentiment trader.com; others include, http://traderfeed.blogspot.com/ and http://www.stocktradersalmanac.com

So I have a zone and setups. What I need is a trigger and initial stop. But before considering that, I need to consider one possible bearish event.

The market came off the top of the Value Area (Zone 1 high); if it breaks the low of the Value Area (Zone 1 low) without first going to the Primary Sell Zone (Zone 2 high to 1576.10 basis cash), the market is telling us that the high probability event is a change in trend from up to down is happening. That assumption will be confirmed with a WPC below 1370.6 basis cash.

So,  if I were to take a position tonight, I’d be looking for signs that the market will rally. We had a ‘test’ day yesterday on the Profile: Neutral Day closing in the middle quadrant. (See Figure 5). So, what we need to see is evidence of strong buying.

 mkt-profile.jpg

FIGURE 5 Market Profile 

Today’s open will be important.

Ideally, we’ll see a Drive or Test Open (see Mind Over Markets) to the upside. This will suggest a rotational day giving traders time to get set at or below today’s developing value area. If we get a test or drive open, we can set the initial stop for the day below the low of today or yesterday. The safest stop would be below the 18-d Value Area. (See Figure 3).

Well there you have it. A Ben Hur of a blog (i.e. a very long blog). All the best and take care out there.

Thanks to Dr Brett Steenbarger

My father was a charismatic character: a champion sportsman who landed in HK to start a job as a lowly accountant and by the time he died, had risen to become the Managing Director with an equity share. He was a well-liked, charismatic character who taught me much: the value of keeping your word, integrity and fair dealing. Among other lessons were his sayings. One of them was:

“The highest praise you can receive is genuine praise from someone you admire and the praise’s content and its form is one you want to be praised for”.

In Brett’s blog (http://traderfeed.blogspot.com/)  yesterday, I felt truly praised:

  • It came from someone I respect and admire.
  • It’s clear that Brett took the trouble to read Nature of Trends. Given that I see Brett as a digital-kinesthetic-auditory personality and that I am a visual-kinesthetic-digital, I suspect the reading was no  easy task. The fact that he did read it is something I deeply appreciate.

So thank you, Brett,  for taking the trouble to read the book and for all the trouble you took to write the review. I am very grateful for what you did.

——————————————————————————-

S&P

Tomorrow I’ll be posting the blog early. I want to cover a full analyis of the S&P and in the process show how Barros Swings, Ray Wave and Market Profile come together to provide a roadmap for a trade.

The Rule of 3

I use the Rule of 3 to smooth out my equity curve and to pyramid safely.

In this post, I’ll explain the first use. It’s important to understand that the Rule of 3 will reduce my bottom line in strongly trending markets. But in markets such as I have been trading recently, it prevents heavy losses and often allows me to exit positions at breakeven or with a small loss.

The Rule as set out in this post is not set in stone. I’ll vary the process to suit my perception of the trade’s risk.

My trader’s timeframe is the 18-day swing (monthly trend). My favourite entry is Negative Development on the First Lower Timeframe i.e. on the 5-d. Figure 1 shows the ideal pattern with a small variation – the trader’s timeframe is the 12-month and the First Lower Timeframe the 13-week.

crude-oil-12-m.jpg

Figure 1 Crude Oil 12-M

Entry is with 3 contracts or multiples thereof. Let’s assume we bought at 60 and had a stop at 52. We exit the first 1/3 as soon as we cover the stop on the remaining 2 contracts. Since our stop is 8 points per contract and we have 2 contracts, we need 16 points. 16 + 60 = 76. We exit the 2nd contract at the Primary Zones, in this case the Primary Sell Zone.

The Primary Sell Zone ranges from 79.35 to 76.54. So I’d exit the contract somewhere in that area. When I exit the second contract, I raise the stop on the third contract to breakeven.

The third contract I hold. This is the contract that gives me a position in case the market breaks out. Most times I’ll get stopped out but the occasions I don’t get stopped out more than makes up for the loss of the profits when I do. Figure 2 shows what happened when the market broke out: at that stage, I have pocketed profits on 2 contracts, holding a third and looking for a retracement to re-enter with 3 contracts and start the process all over again.

crude-oil-12-m-2.jpg

Figure 2 Crude Oil 12-M

I’ll exit the third contract when my trailing stop is hit or there is a change in trend pattern.

What I have described is the optimal Rule of 3. But as I said, I vary the process based on my estimate of the risk of the trade.

In the current ES trade for example, I believed that while an upside breakout was likely, it would not result in a sustained trend. In that situation, I looked to take profits on one contract much earlier. I’d do this by looking for an opportunity to bring the stop on one contract to breakeven. In this way, the second contract need only cover the costs of one stop. I would also take profits on the second contract much earlier and exit the third at the Primary Sell Zone.

Pete Steidlmayer used to have a phrase ‘behavioural parameters: patterns that we can rely on in our trading. Pete taught some behavioural parameters in sideways markets that I find extremely helpful to the Rule of 3…..but that’s a story for another day.

Exit Strategies

A little early today. With CPI due out and the volatility up, I may not get a chance later.

Today I want to talk about exit strategies. Now every newbie will tell you (he may not do it, but you can bet, he’s heard it) that every entry needs to be accompanied by a stop loss. Like most things in trading this is not all it seems to be. Indeed, some fund managers, Ken Fisher, for example, argue against it.

So what is the truth?

One thing is certain, every successful trader/investor I know, and know off, has an exit strategy. If your exit strategy is based on what you can lose, then, you probably won’t be taken out of the game by one loss but you will probably bleed to death. The fact is the market doesn’t care what you can or cannot afford to lose. And as far as stop placement is concerned, neither should you.

First determine what the market has to look like for your trade to be wrong and then work out if you can afford the loss. If you can’t afford the loss, cut down your size. If that still doesn’t do it, by-pass the trade. Willy-nilly placing a stop loss, is just asking the market to take your money.

My exit strategies involve:

  • Price stops: at this price my trade is wrong and it’s a price beyond which I won’t hold a position no matter what.
  • Contextual (strategic stops) exits: what does the market have to look like for me to exit a trade? What does it have to look like for me to remain in the trade? The answers are dependent on the type of setup that got me into the trade. If it’s a Negative Development setup, the market has to prove the trade right after a set time after entry.  I seldom extend the time to exit – though I often exit ahead of time. If it’s a Contraction setup, I am more forgiving as far as holding onto a position that is not moving.
  • Time stops: both Negative Development and Contraction have a time stop.

On Monday, I’ll look at strategies for taking profits.

Trading and Making Money

Two trades in Crude Oil illustrate my approach to making money in the trading game. If there is a Holy Grail in trading, it’s reflected in the formula:

(Avg$Win x WinRate) – (Avg$Loss x Loss Rate) = > $0

In other words, what is important ( taking the win rate into consideration) is for our $ win to be greater than our $ loss.

To maximise the difference, I seek to exit a position BEFORE my stop is hit. So when I say that my trade has an 80% of success, this does not mean that I have an 80% probability of making money. Given that my win rate fluctuates between 47% and 55%, the comment “80% probability of making money” would be sheer nonsense. What the statement means for me: if I exit before my stop is hit, I have an 80% chance of being right. This means that in 80% of the time, if I had not elected for early exit, my stop would have been hit.

Figure 1 shows my results for 2006, my trading year is from September 1 to August 31. My aim is to make around 25% per annum. In 2006 I had a better than average result even though my win rate was slightly below 50%.

2006-results.jpg

Figure 1 2006Results

But early exit comes at a price: there will be times that when exiting a position means I’ll re-enter at a more adverse price. This is what happened with the most recent Crude Oil trades.

Figure 2 shows a classical “Negative Development” buy setup and entry (For Negative Development see previous posts and Nature of Trends). Note that I use CSI’s Perpetual contract for analysis and the appropriate nearest futures month for stops and entry levels.

crude-oil-perpetual.jpg

FIGURE 2 Negative Development

What I expect following Negative Development setup is strong continuation. Instead, we had two inside days. So, on Dec 11, I exited the position with a 0.36% loss. I exited not because the trade had done anything wrong, but because it wasn’t doing what I expected.

I did plan a re-entry on a breakout and this was triggered last night. This second trade can’t lose because I exited 1/3 at last night’s close, moved my stop to breakeven on the second 1/3 and left unchanged the initial stop on the last 1/3.

The early exit cost 0.36%. When you consider the difference between my initial exit and breakout entry, my loss on the trade was 0.42%. But I am happy to wear the loss which with hindsight need not have occurred. But that’s the point: only precognition of last night’s breakout would have kept me in the trade. And since I don’t have that skill, I was happy to bail and happy with the way I managed the trade.

Reviewing Decisions

My assessment last night on what the FED was likely to do was off the mark. I assessed a 67% probability that it would cut the Fed Fund Rate by .25 and the Discount Rate by .50. Instead it cut both by .25.

I always do a post analysis assessment especially if it proves off the mark. I review the decision components to see what I may have done differently, being careful to avoid hindsight bias.

The Decision Tree I constructed rated the probability of rates being raised and remaining unchanged as being so low that I did not spend much time on them. I focused instead on the various scenarios (branches) for a cut in the rates. I pruned the scenarios to three:

  • a) One of the rates would remain unchanged and the other cut by .25
  • b) Both would be cut by .25
  • c) One of the rates would be cut by .25 and the other by .50

The inputs to the branches were:

  1. The state of the US economy;
  2. The need to protect the US $ and
  3. Bernanke’s view of the cause of the 1929 depression.

Reviewing my inputs, it’s clear that I placed too much weight on (3). Given what has been said today about the FED alternative tool box, I was relatively spot on (1) and (2). But where I went wrong was thinking that given Bernanke’s theory of the causes of the 1929 slow down, he would seek to solve the liquidity problem through the Discount Rate. Instead the FED will be seeking alternative means.

What I did get right was the first part of the scenario I had crafted: what would happen if (b) took place. I reasoned that the ES would tank and that this would be followed by two inside days ahead of the CPI on Friday. This is subject to the PPI and Retail Sales not being greatly outside expectations. Once Friday is over, I’ll review the Tree again and see where improvements could have been made.

I do the reviews because I feel it’s important to review one’s decision-making processes. Making decisions that accord closely with reality is as important as reviewing your trading plan etc. Revision of decisions is an important rule for constant and never-ending improvement. And, I feel the review ought to be done not only when we get it wrong but also when we get it right. That way we may identify enabling patterns and limiting ones.

FOMC Rate Decision

There are two services I subscribe to that I find invaluable for the assessment of fundamental data:

  • Shadow Stats: www.shadowstats.com
  • ECRI Light: www.businesscycle.com

Shadow Stats provides an alternative data source to official US statistics and ECRI Light provides reliable leading indicators on the state of the US economic growth and inflation. The two together provide data on whether or not my ideas about ‘long-term’ perspective are being reflected in the economy.

I tend to rely more on the ECRI report than Shadow Stats and will lean to the former if there is a conflict. But right now they both agree: the US economy sucks. Despite the rosy picture provided by Non-Farm payrolls, the latest Shadow Stats demonstrates why the official figures are an illusion.

The ECRI published Friday Dec 7 2007 for the week ending Nov 30, shows US economic growth at a 5-year low and inflation down a tad. The ECRI indicators have about a 3-months lead time. So, I’ll know if my idea – that the massive increase in the money supply in August through to November 2007 will lead to a rise in inflation despite the weak economy – will be correct about 3-months before there is pressure on the FEDS to raise rates. But that’s for the future.

For today, the true state of the economy is the focus. Tuesday 2:00 PM EST, the FEDS will announce their decision. The worry for me is I am in the majority camp – I think there is little chance there will not be a .25 cut in the Fed Fund Rate and in the Discount Rate. But given the state of the economy, the sub-prime liquidity crunch, and Bernanke’s ideas of the causes of the 1929 Depression, I rate the probability of a hike above .25 around 67%.

If it were just the matter of the economy, I’d rate the probability much higher. However in the near term, the FEDS have the US$ to worry about. I would hesitate a guess that despite the jaw-boning, they would not be averse to a sliding US$ as a measure against recession. However, what they don’t want to see is a US$ rout. And that is what they are likely to get if they decreased rates in the Fed Fund Rate and the Discount rate by 0.50. In addition, the decrease in rates in the Fed Fund Rate has done little to assuage the liquidity problem.

So, for all these reasons, I rate the probability of about 67% for a decrease of .50 in the Discount Rate and a .25 cut in the Fed Fund Rate. Such a move at least provides some chance the US$ decline will not turn into a bearish stampede.

What do I see happening if such an eventuality occurs? Well, I’d expect the ES to rally strongly and take out the current highs before year-end. We’re only 70 points or so away, so that’s not hard to imagine.

I’d expect the US$ to drop strongly over the next week but do expect to see some buying come in. December is seasonally a strong month for the US$ and so far it has held up well.

As always, whatever your position (long or short), in whatever instrument (Gold, Interest Rates, Stocks and Stock Indices etc), I recommend you create some worse case scenarios in case the FEDS do something unexpected and/or the market reacts in a way contrary to your expectations. Preparation is my key to preventing the ‘rat brain’ seizing control and thus turning a well-planned loss into a catastrophic one.

Zones in a Sideways Market

In “Negative development As A Setup”, I attached the chart in Figure 1.

sp21.jpg

FIGURE 1 S&P Zones in a Sideways Market

I have been asked to explain how I use them and how they are calculated. So here it is.

Zones 1,2, and 3 in Figure 1 represent 1st, 2nd and 3rd standard deviations of the boundaries of congestion, using Market Profile TPOs. For a full description of the process, go: http://www.cbot.com/cbot/pub/page/0,3181,1184,00.html where you can download the CBOT Market Profile Manual.

For those that would prefer a quick and dirty calculation: Take the range of the boundaries of congestion and divide it by 8. The sell zone (Primary Sell Zone) is “High -1/8 Range”; the buy zone (Primary Buy Zone) is “Low + 1/8 Range”. These zones represent the 3rd standard deviations.

The 1st standard deviation is represented by the 33% and 67% retracement levels of the boundaries of congestion. The 50% retracement corresponds to the Profile’s Point of Control.

Note that the retracement levels are approximations only of the statistical zones. But they are close enough approximations to use.

So that’s the way I calculate the zones – the full method as well as the ‘shortcut’.

Let’s turn to the way I use the zones.

I use the Primary Buy and Primary Sell Zones as areas where I initiate and liquidate positions. I also use the zones to tell me if:

  1. The sideways market are likely to be coming to an end; and
  2. I use them to manage a trade.

Let’s take point (1).

The Primary Sell and Primary Buy Zones are rejection areas. If a sideways market is to continue, we should not see buying conviction bars in the Primary Sell Zone (High -1/8 R) and selling conviction bars in the Primary Buy Zones. If we see two consecutive conviction buying closes in the Primary Sell Zone, we have a clue that the Primary Sell Zone will give way. You reverse the process for the Primary Buy Zone.

An example of point (2):

Let’s assume the market comes off the Primary Buy Zone (Low + 1/8R). The market then has two consecutive closes above opposing 1st standard deviation (in Figure 1, Zone 1 High, 1513.60). If the market now closes below the Zone 1 Low (Figure 1, 1427.60) without first entering the Primary Sell Zone, we can expect the boundary of congestion low to give way.

This is the great thing about the Market Profile: it’s full of these little gems that help a trader manage his trades.

Preparing for Figures in the S&Ps

Non-Farm payrolls tonight. The forecast range is +110,000 to – 10,000 and the mean is around 85,000.00. The mean was around 65,000 to 70,000 until the ADP employment number last night. With that coming in at 189,000 (against the expected 50,000), the expectations have edged up.

In this post, I’d like to show the process by which I prepare for figure night. Figure nights have been my bane. It’s the time my ‘rat brain’ (using Dr. Janice Dorn’s label) is most likely to exert an influence. To counter this, I spend a lot of time preparing my responses to the figures; in addition, I make a commitment to adhere to the pre-figure plan. In short, it’s one of the few occasions, I don’t give my intuition a look-in; at least not until at least three or more hours have passed.

The first thing I do is the normal analysis – as if there are no figures coming out.

Right now, I am long with expectations that the market will make a new high. I intend to cover the longs on new highs and stand aside after that. I outlined the fundamental reasons why I think the next high will mark a 2-year high in the post “Fundamental or Technical Trader?’; add to this the position of the market on the Ray Wave structure and it makes holding past new highs a high risk trade

In the meantime, using a variation of the Rule of 3, I have closed out 1/3 of my open positions, brought the stop to breakeven on 1/3, and for the remaining 1/3, I have my stops in their initial location. The profits I took will cover the loss on the 1/3 initial stop so I am on a ‘no-loss of capital’ situation.

So the above is where I stand at the end of the normal analysis.

After the usual analysis, I prepare my responses to the figures. The attached Decision Tree shows the probabilities as I see them and they show scenarios that are too close to call.

The events I considered:

  1. The ADP numbers have a terrible record for forecasting the Non-Farm numbers but they were so skewed, we may see a higher number than expected.
  2. On the other hand, John Williams’s excellent site had this to say: (http://www.shadowstats.com/cgi-bin/sgs):

“Employment Numbers May Play Role. This Friday’s employment report could be used to decide or at least to try selling any forthcoming rate action or lack of same. Fundamentally, the numbers should be horrible; October help-wanted advertising sank anew, while jobless claims continued to rise.

So the first thing I decided was I wouldn’t be doing anything in the ES immediately after the Figures. If they come out within the range +100k to +10K, I’ll call it a night for the ES. If they come in at either extreme, I’ll be looking for a sell-off and then signs that the sell-off has failed.

The sign I’ll be looking for is an ‘open-gap’ down at the open, followed by a failure to close 50% of the gap in the first 60 minutes. The hourly close would need to be at least around the 50% of the hourly range and preferably at or below the lower 33%. I’d then look for an upside breakout of the hourly range.

On the breakout, I’d need to see Market Delta confirming breakout volume.

decision_tree.jpg

DECISION TREE