Short Term Market / Sector Direction Observations

[mw]

Shorter term observations / swing trading ideas and other directional ideas. I intend this thread to be a storehouse for on-going thinking; perhaps those interested in market timing can add some value here and conclusions which seem to be longer term in nature can be posted in the Market Direction Thread.

[mw]

Sorry for the very tall chart.

Yesterday I posted this, identifying a number of targets for a market pull back. Rationale for considering a decline? Price action on Friday and Monday (US) set up two topping patterns (candles on the daily bars). In addition to price action alone there was the backdrop of VIX hitting a months-long low, and a failure of financials (US) to participate in the attempt to move higher out of the recent range.

First steps first, price needs to break Friday's low to even initiate a downswing on these charts (as of writing, both the Nasdaq Composite index and US financials have).

Still, one would not expect a major market direction change unless last Monday's low is taken out decisively. Any pull back here has merely brought price back into the chop, although using larger time frame intraday charts (15 minute through 65 or 130 minute) we ought to be able to detect an upper range trading failure and use that as another justification for defensive action.

[mw]

Dow Theory folks might also appreciate that the TRANS index has so far failed a test of top on the weekly / monthly charts:



And a spike top and immediate price reversal back into the uppermost trading range formed on the daily over the past couple of weeks:



A failed test of top isn't the same thing as a market reversal - one can't know that has occurred until later. But it is a reason to adopt a certain amount of defensiveness, particularly if one is invested in the US and/or transportation sector. Specific to this case, this sector is watched by many.

Price will have to reach below 5163 on the TRANS index before we can effectively say that the uppermost trading range (on the daily chart) has been broken, and so we shall have to keep on watching over the next week.

[mw]

Now with the overall scenario laid out, the next task for us is to determine where the upper most trading ranges are (or in the case of the financials, the most recent trading range).



12715 on the Dow marks the lower boundary of its uppermost range. 1380 on the SPX. 2492 on COMPQ (which had developed a smaller, tighter, trading range in recent days).

When price starts to *trend* (lower swing highs, lower swing lows) below the upper most major trading range on a 65 minute intraday chart, in conjunction with the daily and other longer term charts illustrating failed tests of tops, then at that point one can comfortably say:

The up trend in the market has reversed direction and is now a down trend until proven otherwise.

And at that point this note should be moved over to the longer term market direction thread. To this point we are merely identifying the technical preconditions for a significant trend reversal.

For now it would not surprise me to see some bounce into overhead resistance. Whether that proves to be a good buying opportunity or the last opportunity to sell at higher prices for a while is yet to be seen. One can't know for certain until after the fact, but there are enough clues to warrant caution - that is the point I am / I have been trying to stress as of late.

As folks here know, I've been getting defensive, and in the case of financials, overweight in the short direction, over the past little while, anticipating a market turn. My portfolio is essentially long energy (but lighter than I have been in years), and short short short financials (US and heavier weight in Canada... which has certainly lagged the US).

[mw]

Time for an update on US market direction, as the Dow 30 comes in for a touch down at the first major target, the range lows. This is a busy chart:



Should price hold within the range, all we can say is that the longer term bounce off March lows remains intact but the market is in consolidation. Periods of consolidation I always view as indecision - buyers and sellers are more or less balanced out.

Should price not hold the range lows, then the approach I use says that change of trend from UP to DOWN has occurred until proven otherwise (which will require price back up in the range).

In the early days (or hours or minutes) following a key range break (up or down) market participants are trying to determine if its a buying or selling opportunity. Rarely does a market break straight down or shoot straight up. For the observant there will be good and safer opportunities to position long or short. The methodology I use makes the market take me into a trade, proving where it wants to go, rather than guess work.

Should price break lower and fail to regain the range, then we can look for either Fibonacci retracement percentages for potential support or prior tall up bars. I've marked some likely targets - as price moves lower the chance of a full retracement of course becomes increasingly likely.

How do I use this?

Knowing that a potential change of trend, or a reconfirmation of support is occurring before our eyes, I use this opportunity to bring stops up on market-following names to take profits on remaining position balances, but I am also looking for names which offer long-side setups in case the market finds its legs. On the short side if price fails to hold then I know we have the makings of a new downtrend and will look for retracements (bounces) in that down trend to sell short. Others might employ the first bounce or two to lighten up long positions.

The key piece of information is that once a trading range sets up one can easily employ that information to detect reversals and continuation, and if one can do that, then you can avoid going long in declining markets or selling before a market reversal.

Bottom line at this point: suspense. And yes that might continue for a few days if no catalyst for movement appears.

[mw]

Step one complete. I like to think of this as a presumptive change of trend. We've got a test of top, failed. A test of range bottom, currently failed. A lower swing low in place, check. All that is left is a relatively minor bounce (could even go part way into the range above, thats a truly diabolical yet not uncommon scenario) and then another round of selling to set a lower swing high. A failure to hold today's lows following that bounce and voila, a change of trend on the daily charts.

If I were not already net short the market I'd be looking at bounces to locate selling points, but that's a topic for another day.

[mw]

US markets continue to sell off; Dow 30 index has confirmed a change in trend from up to down on the daily charts now. Price action of the past week has undone all that which was achieved over the 4 week period prior. Price has almost retreated 50% of the entire mid-March to May rally. That alone may bring buyers in but I won't be one of them, not without a clearly cathartic day or period of torment.

Just days ago bullish market sentiment had become almost pervasive. The worst is behind us, they said. Nothing but good times ahead. Emerging markets are picking up the slack. Aren't earnings resilient / great / outstanding?



Why then has price stalled at a test of the swing high?

There isn't nearly enough fear out there for the current lows to become a bottom for this decline; that and experience suggest that bounces from here will be relatively minor and/or contained underneath the uppermost trading range (bottom edge near 12720). Experienced traders will be looking to dump long exposure on bounces or use any lift in the market to locate short positions at higher prices.

Nothing to do here except hold short positions already open and look for bounces to sell.

In the longer view, should price retrace all of the gains since March, its likely that an even steeper decline is in the cards for this year.

[kcowan]

Although I have nothing to add, I am enjoying you commentary immensely. Please keep it up.

[gyrfalcon]

Although I have nothing to add, I am enjoying you commentary immensely. Please keep it up.

Agreed. The style is a bit telegraphic, which therefore requires some exposure to the lingo, but the content is comparable to the stuff folks get paid to write.

[mw]

Thank you both sincerely for the feedback; its a little quiet in here, akin to "thinking out loud", and I wondered if anyone was finding anything of use or interest. Pointed noted about lingo, I will try to go into some depth and develop along the way a glossary of terms.

In the end I hope what flushes out of this out-loud thinking is tips or techniques which could have applicability to the longer term diversify and allocate group of investors, as well as anyone with even a passing interest in market timing. Certainly questions, observations or contrary opinions would be a welcome input along the way.

Today/Sunday I plan on posting a number of charts and commentary which should go some distance to explaining why I've gone net short in all our accounts, what I expect, and how I'll cope with the fallout if time proves the decision to be wrong.

[DenisD]

I enjoy your comments too. I'm wondering why you use Dow charts rather than the broader S&P 500 ones?

[gyrfalcon]

... Point noted about lingo, I will try to go into some depth and develop along the way a glossary of terms.

In the end I hope what flushes out of this out-loud thinking is tips or techniques which could have applicability to the longer term diversify and allocate group of investors, as well as anyone with even a passing interest in market timing. ...

Anyone who really wants to move ahead on their understanding could start here:

http://stockcharts.com/school/doku.php?id=chart_school

I use stockcharts for all TA that I do (much more limited than yours). I would recommend the "Sharpcharts" & their numerous related features. HINT: Cdn stocks are ABC.TO, Cdn Trusts are ABC/UN.TO & US stocks are simply ABC. gyr.

http://stockcharts.com/charts/

[mw]

The Trend Is Your Friend
Until It Ends

Market timers, swing-traders, momentum players, and long term equity investors, all have a common goal: they expect to gain capital appreciation over time. The principal difference between these classes of investors is the time frame for investment or trade is different.

Realizing a profit through capital appreciation requires price gains which result from participating in an uptrend. This is the case regardless of time frame or investor type, whether the time frame they operate in be measured in minutes, days or years. Understanding the mechanics of a trend, how to detect an emerging trend and most importantly how to determine when a trend has ended, can help any investor class improve their performance.

The following chart is going to serve more than one purpose however first we shall use it to establish a few basic definitions and principles, some of which will already be familiar.


Figure 1: US Transportation Index, Monthly time frame

What is a trend? Put simply a rising trend is a series of higher highs and higher lows. We can be somewhat more precise and state that a time series chart of an equity (or bond or household prices or any price-based object or commodity) which shows a set of higher *swing* highs and higher *swing* lows is in an uptrend. Naturally a set of lower swing highs and lower swing lows is a down trend. Importantly we must note that what is trending in one time frame may not be trending, or may be trending in an opposite direction, in smaller or larger time frames.

What we will not look at. Among the technical analysis community there is a wide variety of techniques used to locate trending markets and entry or exit points within each. We are not going to discuss at any time the many and varied indicators commonly in use - most all of which are re-interpretations of price. An enormous amount of effort is invested in the development of these "tools" in the search for the holy grail--my opinion is that most of this work is wasted effort. The techniques that I use and will share are based on simple price observation and work whether one is charting on the latest computer hardware or doing it by hand as they did back in the days of Dow.

Identifying trends. How to we locate trends already in progress?


Figure 2, constructing a proper trend line

Remembering that an uptrend is a series of higher swing highs and lows, we can mark up our chart to highlight a uptrend using a simple trend line drawing technique. We draw a line starting at the lowest swing low, passing the line through and beyond the bottom of the highest swing low present in the series prior to the current highest swing high. The one proviso we must follow as a rule is that the line shall not be intersected by any price until after the highest high in the series has been made.

Essentially we are looking to delineate the easy to spot trend, and as the trend progresses, it will become obvious that redrawing the line will become necessary. More on that later.

Detecting the start and end of a trend. How can we detect when a trend has ended, as opposed to having merely stalled? Conversely, how can we detect when a trend has started?

Eventually price will make a significant move below our evolving trend line. The simple rule to remember is that until a higher high has been set, our line remains rooted at the lowest swing low immediately prior to the highest high on the chart.

Anatomy of a trend change: Step 1, the uppermost trading range


Figure 3 - the development of a trading range

Corrections such as shown in Figure 3 occur frequently over the course of a bull market. How can we identify whether the bull is merely pausing as opposed to reversing? The patient trader or investor will need to wait. Generally a correcting market will attempt some sort of rally and at that moment we can use the points of reference laid out on the chart to mechanically determine if the market is reversing or carrying on following its correction or "pause" phase.

Two more lines are in order.


Figure 4 - identifying the uppermost range

Here we are marking out the current known limits of the uppermost trading range which has developed following the perhaps as yet intact long term uptrend in the Transportation Index.

All we know at this point is that price is moving net sideways - the overall trend - at this point almost five years old - has not yet reversed. Long periods of consolidation are not abnormal but a deep correction such as portrayed is an indication that we really need to be paying attention to the aging bull run.

While price remains in the trading range we can't venture an opinion as to whether it is likely to continue or reverse, but certain milestones will help us make that determination eventually.

Anatomy of a trend change: Step 2, the failed test of top

Figure 5 - a "2B" test of top

Of all the events which we can observe from price in an uptrend, none carries more significance than a retest of a prior trading range high. A failure of price to carry on is a signal which many market participants regardless of their style will take note. Price is something we all understand and such tests are part of the lexicon of investing life whether you look at charts or not. 52 week high lists in the news paper. TV faces inform us that the market is "Trading near the highs of the day".

Figure five shows the current chart of the US Transportation Index. We see that - so far this month - price has made an attempt to break the prior range highs and has immediately reversed. This marginal new high and reversal is known by some technical analysts as a "2B test of top", classified as such by Victor Sperandeo in his book Methods of a Wall Street Master. Incidentally he is the first I've ever seen to outline a sane methodology for drawing trendlines and his method is what I have adopted and illustrate in this article.

At some future point I'd like to discuss emotional response to price but a short note on this chart is warranted. When price makes a marginal new high there are a range of emotional responses felt by various market participants. Previous highs are frequently the location of buy and sell stops, for those who want to get long, or want to end a long trade, as well as those who were short but need to bail, or those who want to start a short trade. When price reverses you can be sure that most participants are going to be off-side, and this creates some initial thrust or fuel for the move down.

Such it is and has always been in the market. Ratios and expectations mean something, but do not mean as much as losing or gaining money, and certainly ratios and logic are less important when price is at extremes. Emotional response is at its highest when markets are rocketing or reaching new highs, and at its lowest when hitting new lows with increasing speed.

Is the failed test of top a sign of reversal in the market?

No, and yes. No, the failed test of top does not in fact reverse our long term uptrend, because no series of lower swing highs and lower swing lows is yet in place. Thankfully we have a completely reliable totally mechanical way of identifying trends.

But yes, it is a big red flag, particularly when looking at charts in larger time frames (weekly or monthly charts).

What signals the end of the trend?

Refer back to Figure 4 and consider the following statement:

A trend will have reversed from up to down when the following occur:

1. Price moves below the uppermost trading range lows
2. Price initiates a series of lower swing lows and lower swing highs.

Clearly by the time a new downtrend is in place on the monthly chart, a significant portion of the gains from the prior uptrend will have been lost. Our solution? At key inflection points (tests of tops and especially at the point of departure (not shown yet on any chart) of the range lows, dial down the analysis by at least one if not two time frames to either the weekly or daily charts.

Myself, I use the test of top as a presumptive change of trend in the works, and take a failure of a test of top in larger time frames (weekly or monthly charts) as an instruction to lighten up my long positions or perhaps exit the market entirely or add short exposure. When reviewing markets off the larger time frames there are fewer false alarms and if the market should manage to find its footing, there will be ample time to change stance and rebuild long positions.

Others with very long time horizons might wish to wait for the return to the range lows before taking decisive action. After all, it may be that the market decides to move sideways for many months or perhaps even years, rather than initiate a new downtrend on the monthly chart. While that sounds like a reasonable position to take, bear in mind that a sideways moving market on the monthly chart may have 20 - 30% swings in price movement for many years with no guarantees as to ultimate outcome.

Finally the double top on a monthly chart is not something which comes along very frequently. With regard to the transportation index we've been reviewing, the last such case was the range which was defined first by the sharp price declines in the 1998 financial crisis; the Transportation index rallied to a marginal new high in June 1999; the index fell over subsequent years to a decline of 49.5%.

[mw]

With some basic terminology under our belts - trends, tests of tops, ranges - and an approach to determining trend reversals using price departure from ranges at the extremes, lets move on and talk very briefly about sentiment and then specifically about the current market environment.

While I do trade and monitor the S&P 500, I frequently use the Dow 30 for examples because the index is a useful sentiment indicator all on its own. Its the most quoted index on the planet. Its the only index where people partied when it crossed a big round number, 10,000. The fact that price later recrossed, and recrossed it again and again somehow has managed to escape the media but perhaps we'll have another Dow 10K party before this year is up.

Thus its useful to follow the Dow 30 merely because so many do. In addition there remain quite an active cadre of Dow Theorists, so we want to keep our eye on both the Dow 30 Index ($INDU) and the Transports index ($TRAN).


This chart is becoming familiar:


Figure 1, Transports index, monthly time frame

I wanted to make a comment about the spike top, still in progress of forming, on the as yet not completed monthly bar. Spikes - where price opens and closes near the low of the bar, and in between the open and close price heads up and right back down again - represent extreme shifts in sentiment. Hope (of new highs and a breakout) and Terror (failure, reversal, time to take profits) all in one bar - and in this case a bar spanning *an entire month* can't be ignored.

When spikes are found at important tests of tops, particularly in larger time frame charts such as the weekly or monthly time frame, we really ought to be on high alert for a potential market correction or worse.

If one reviews such spikes on monthly index charts (in the example above the spike bars are coloured red using a simple formula) you can see that price declined for at least the following month, or headed/remained lower for a period of some months. Tests of tops up the ante. We'll come back to that shortly but first lets look at the Transports and Industrials together.

Dow Theory - Buy or Sell Signal?
Recently I recall reading an alert from some news service or newsletter writer that the Transports were signaling a Dow theory buy confirmation. I sure don't see it. In a nutshell Dow theory suggests that the Dow Industrials and Transports ought to be strongly correlated to confirm a trend; since the fall of 2007 nothing could be further from the truth.


Figure 2 - Dow / Transports overlay, weekly time frame

But Dow theory folks can (probably) rejoice as of last week, for if the market downturn thesis I have proposed plays out, last week's failed test of top in the Transports will have Transports and Industrials both moving in the same direction - down - fulfilling one of the requirements for a Dow theory sell signal.

Of course the Industrials could sort out the divergence by firming up and rallying to new highs. That outcome just doesn't seem likely in the intermediate term.

Lets dive in deeper and look at the daily charts.


Figure 3 - Transports - daily time frame

Within the developing spike bar on the monthly chart we see a trading range has developed on the daily chart and price has reversed to the lower end of that range. Technically this sector has not broken down yet - the worst we can say about it, in this time frame, is that it is in a period of consolidation. If price breaks down out of this range then we'll call it a presumptive reversal of trend until we can see evidence which confirms the downtrend - lower swing highs and lows on a time frame one down from this (130M or 65M charts).


Figure 4 - Dow - daily time frame

The Dow 30, unlike the transports, has already provided sufficient confirmation to rate recent price action as a highly probable trend reversal. Since the failed test of top the Dow 30 has displayed price action characteristic of a reversal. Provided price does not return into the range (quickly) we expect to see lower swing highs and lower swing lows on the daily charts. These are already found on 65M intraday charts. Will it bounce some? Perhaps. Might be a good time to lighten up.

(The SP500 has similar price action but has only broken the lows of its uppermost trading range by what I would consider a marginal amount. The slightly stronger relative strength of "tech" (adding to SPX and Nasdaq Composite) is perhaps the strongest contrary argument to my market decline thesis. Unfortunately they aren't all that strong if one steps back, for in the bigger picture the weekly SPX is more or less a mirror image of the Dow 30.


Figure 5 - Dow, The Big Scary Head and Shoulders Top

Edwards and Magee wrote about a price target forecasting approach which uses price extension - essentially taking a linear measurement of a range, and applying that from the point of departure (a "breakout") to obtain a target. In fact I've found this to be a fairly useful technique and use it in all time frames to develop price guides. I don't particularly like doing the measurement off a Head and Shoulders pattern because the extension typically covers an unbelievably large range - for example one can make the case for price declines in the Dow to near 10,000. I don't have trouble with that myself but many would and it causes them to discount such potential out of hand.

What of other markets? Won't | insert emerging market here | "save us"?

Short answer: No, I don't think so. A goodly number of international markets are mirroring the basic price action of the Dow and SP500. Those that aren't (Canada and others we'll cover in time) have energy as a strong component of their indexes or have other issues at work. Brasil / Latin America is an interesting case, but I do see some potential for a buying exhaustion top in place there. Still these markets (and Canada too, off fewer and fewer leaders) could continue to deliver gains for a surprising length of time even as Rome burns. Remember Nortel-lifted TSX back in 2000? That didn't turn out well, did it?

VWO, Vanguard's Emerging Markets ETF which follows the MSCI Emerging Markets Index, appears to be joining other senior markets in reversing the run up from the spring lows. While it certainly has performed better on a relative basis, such gains can evaporate as price returns to the mean / joins the rest of the crowd. I've not got strong feelings on this other than to state that if price can not find support between 100 - 101.50 I would expect- the gap near 97 to be filled and finally if 95 - representing a 62% retracement of the spring rally - can not hold then a complete retracement to near 87 becomes more likely than not.

Despite all the talk of late about "decoupling" from the U.S., when push comes to shove and sellers are hyper active in the U.S. I don't see much decoupling happening - everyone moves down. Witness Asia over the past 8 sessions.

Summary

In posts that are far too long for the content within I've laid out some basic definitions and a description of the trend and tend-end detection technique which I use in time frames running from second charts through weekly, monthly and quarterly charts.

Major U.S. markets are falling over one by one, confirming a reversal in trend. Bounces from here over the next few sessions ought to be relatively weak or fail to hold up by the end of session. Provided price does not close well within the uppermost trading range which we have marked, there is no other conclusion I can come to other than to simply state that a trend reversal from up to down on the daily charts has occurred. Target one for most markets and sectors I have looked at is the spring lows.

[northbeach]

I am enjoying this thread.

My brain certainly does not work like yours.

Thanks

[mw]

Figure 1 - TSX Weekly chart

Marked tenative - price has to dip further and start to trend down in smaller time frames before we call it a confirmed failed test of top.

The problem here is that one simply can't know in advance whether its going to fail and head much lower or not. The technical trader doesn't argue with price therefore this is a big heads up - if price can't emerge higher then some traders are going to assume a trend reversal is in place.

Technically the major trend isn't reversed until price breaks through the lower boundary of the uppermost trading range in the time frame of reference (this chart is a weekly timeframe).

Waiting for that confirmation would represent very large losses so we have to move to smaller time frames to get confirmation rather than give up that much ground in the portfolio. With my longer term holdings over the past week I am looking for reasons to hold and reasons to sell short or gain short protection/hedges and considering each on their own merit.

[mw]

Dow, SP500, US Financials and Nasdaq all closed out the week lower than last week. With the exception of Nasdaq, major US markets have been unable to rally into the upper end of last week's range. "Bear flags" - minor retracements within a more dominant move - exist on Dow, SP500, Financial sector. Tech is trading within a range.



Can Tech lead the overall market higher? At this point I'd have to say no, as its less likely that tech can stage a recovery for all when Financials is such a drag on the market. US traders/investors looking for clues would do well to watch Nasdaq and Financials for direction.

Briefly today we saw Financials assume leadership - at the open - but that was completely reversed during the day. Perhaps they can try again next week.

Now that the month is done, lets have another look at the very large time frame charts. I've added bar highlighting to show May-June pairs as we explore the theme Sell in May... Walk Away.

First a recap - The Dow and Financials sector closed lower than the prior month; S&P 500 closed marginally higher thanks to tech weighting (and energy no doubt) in index but with a narrow range overall compared to recent months.



Personally I do not consider the sell in May folklore to be a workable pattern all on its own but certainly there is less liquidity / trading activity at times during the summer so moves may be exaggerated once they get started.

Of the broader indexes both the Dow and SPX failed to push and close higher in the marked out trading ranges pulled from the daily time frame charts.

Looking at the charts for similar weak periods for parallels with price action this year to date one can't help but look at May/June 2001 on the S&P500 (circled) or May/June 2002.

The parallels are reasonably similar from a price action perspective. In Spring 2001 the broader market was suffering due to one bubble popping; in Spring 2008 the broader market is suffering due to another, more important, bubble popping - financials re housing and other credit related issues.

In Spring 2001 the market rallied briefly ending in a doji or "pause/indecision" bar in May and headed lower (for many months to come). In some respects financials spring 2008 is not unlike the tech stock bounce in Spring 2001. While one could make a comparison with April/May/June 2003 - the start of a new uptrend arose from there - the consumer is in a much different place today than back in early 2003.

The only thing we can state categorically about direction is that the down trend in the very large time frame remains intact, with lower swing highs - so far - and certainly lower swing lows in place. Should price regain the shaded grey areas I would want to re-examine this thesis and would look for long side entry points again with more vigor.

There are not many parallels in time during the past 14 years to this period; the tech/financials bubble comparison is the closest we can come up with in this era.

Personally I'm still short financials and started a Dow short yesterday and added to it today. I'll be looking to see if price can break Friday's lows early next week and if not I'll probably stand aside and may even choose to take some profits out of my May 2 and May 5 US financials short positions. I'm interested in being patient here because my first entry is within spitting distance of the May highs and there is high potential to see the sector reach new lows this year. If that transpires we should see the entire market lower.

So my plan for next week, if price sets a new weekly low in financials I will hold the entire position unless a very violent reversal up comes about.

I would not be too surprised to see prices merely meander sideways for a while either but am hopeful the narrowing ranges of the past few days lead to volatility expansion (volatility implies no direction just.. volatility) once more.

It was much easier to get long in a big way back in the depths of March than it is today - when price is moving fast its always easier. To date I've not had any regrets about lightening up and getting short certain markets (although if the post release rally in Canadian banks holds up I'll be somewhat annoyed at the dead capital in that one trade grouping).

edit: no changes made, editing wrong message

[adrian2]

Dow, SP500, US Financials and Nasdaq all closed out the week lower than last week. With the exception of Nasdaq, major US markets have been unable to rally into the upper end of last week's range.

Where are you getting your info from?

Other news outlets, including radio, beg to differ:
http://biz.yahoo.com/ap/080530/wall_street.html?.v=46

All three indexes finished higher for the week, recovering from the previous week's sharp losses. The dollar stabilized and oil prices pulled back from record highs during the past four sessions, giving investors some relief as they parsed data suggesting that the economy is weak but not technically in recession.

The Dow rose 1.27 percent, the S&P 500 gained 1.78 percent and the Nasdaq picked up 3.19 percent.

The financials index ended today at $82.56, last Friday it was $81.70

[mw]

Actually you are absolutely right; I suffered from a brain lapse there when I wrote the sentence. What I'd meant to write is that all markets closed lower than last week's high and open, and that all bars except for the Nasdaq were "down" bars.

The distinctions I meant to make (thinking but not writing clearly) is not inconsequential, and I'll post something on that shortly which will go some ways to explaining what I was thinking while my fingers merrily typed a confusing sentence away.

[mw]

More Terminology - Up, Down, Inside, Outside Bars

One aspect of the analysis I perform on markets involves looking at absolute movement as opposed to net movement. A chart will help:


Here I've applied some code which colours bars based on their absolute movement relative to the bar before. I've taken off the open and close ticks because they are not relevant to this discussion at this point.

If a bar has a higher high and higher low than the prior bar, regardless of the close, its an Up bar. The reverse is true. A bar contained within the range of the prior bar is an inside bar - a narrow range inside bar can be seen as a point of indecision. Useful creatures we ought to talk about later at some point. An outside bar might also be considered a point of indecision; some might rightly consider it a time of violence. We'll talk about this more later too.

The important thing to remember is we are talking about absolute movement without any regard to the closing or opening prices.

When I talk about a "down" bar, as I had without realizing it in a post earlier today, I'm referring to a bar which has a lower low and lower high than the bar prior, regardless of closing price. This way of looking at bars is so ingrained in my thinking that I don't always realize that many do not know what I am talking about.

Same chart, closing ticks on:


Seeing the closing ticks will often leave a much different impression, or at least should. A close in the lower range of a bar suggests price is likely to travel lower in the next bar, and vice versa. Yet one ought not to forget that the price bar itself indicates the range price was able to trade through and this will tend to repeat itself, especially in larger time frame charts such as monthly time charts.

Lets look at two time frames up:


In this monthly chart I've drawn in the swings based on a mechanical formula. The last swing is of course tentative as its still in play. You'll notice that within a swing bars headed in the same direction (up or down) will be the dominant bars thus an upswing will have more up bars, a downswing the reverse.

One can exploit that bit of seemingly obvious knowledge. Lets move back down one time frame to the weekly chart:



Knowing that within a swing there will as a rule be more bars that point in the same direction as the swing than not, one can make a habit of looking for outliers - for example in the above chart I've singled out instances where a SINGLE UP bar showed up within a still-intact DOWN swing.

We can play a number of scenarios out, and have a chance to earn some money speculating on direction. A single up bar in a down swing might be the start of a new up swing - without going into the details of the mechanical system I use, another bar is generally required to confirm. If for other reasons our analysis tells us this market is likely to head down further, an up bar in a down swing gives us a perfect opportunity to locate a short sell triggered only if price moves down below the UP bar in the next bar.

The reverse would be true for long-side entries; if we have a down bar in an up swing (ideally an upswing as part of an up trend), we can look to go long above that bar.

What this allows us to do is enter a trade only if price proves itself to us by heading in the direction that will stop us into a trade. Instead of catching a falling knife in this case we are selling short that falling knife.

Lets dial back up one more time frame to the monthly chart and finish this off:


Here we have what I consider a relatively weak bounce given that price has largely remained below the 38% retracement of the long term decline. Price closed in May near the bottom end of the range, and lower than the April close. Its likely, but not a given, that price may break the May lows early in June which will be an occasion to enter into a sell short trade with a target of the spring lows. I personally do not trade off the monthly chart - my short entry is much higher, a counter trend trade (on the daily chart) entered in early May - but I do use the larger time frames to ensure I am trading in the right direction and/or to filter out noise.

We can't know for certain if price will head back to the lows and beyond, so its rather nice to let price prove itself that it is weak; by not entering a trade now based on a supposition price will move lower we've helped eliminate some of the direction-risk. If CM goes up directly in the next bar (month) our sell short order is never triggered.

Incidentally the blue arrow on this last chart indicates where the current and still valid up-swing will revert to being a down swing again.

Getting back to the error in my earlier post, "down" for the week was my brain looking at a chart like the Dow 30 weekly:



... which, like the SPX and XLF display a down *bar*. Sloppy proofreading on my part failed to catch that I said "closed out the week lower".

[DenisD]

mechanical system I use

Just wondering how much of your trading is controlled by mechanical systems and how much is discretionary? Or, do you use discretion with your mechanical systems?

[mw]

Good question: All of my trading is discretionary; when I refer to "mechanical system" I'm referring to the method by which I determine if a market/equity/symbol is in an upswing or a downswing in a given time frame. While I did use the term mechanical repeatedly in context with the determination of trends and swings, I realize the term has another much different meaning in context with trading systems and ought to have been very explicit that I am not talking about the latter.

I've done some exploratory work (quite a bit over the years) with truly mechanical trading systems but I do not trade them. What I will do with such tools is use them to look for very particular conditions in markets that I follow. A list of buyable retracements in energy, for example, is a handy thing to have available each morning.

The actual decision to invest or trade always remains a discretionary act.

[mw]

Sadly I had to close out my NDX short (noted yesterday in the What did you buy thread -- using QID) as in the range yesterday afternoon there wasn't enough profit buffer to take a chance with payrolls data being somehow more positive than expected or with traders ignoring it, as they did with oil and retail sales details, neither of which were positive news yesterday.



So I've entered a short here and will re-enter a short via QID if my break even stop is hit today.

With Dow again near its month lows and threatening to break those lows the chance of a whoosh down are rather high; the dominant trend in the broad market is down and the NDX is fighting that thus if it reverts to the mean it will play catch up to some degree which is why I want it in my short portfolio.

If it means I have to find another entry point today I will; not unless the Dow/SPX can reverse all of today's losses will I again consider that the market may be trying to reverse the dominant down trend which remains in place on the weekly and monthly charts. Its going to be a long day

[lilbit]

Hi mw, I've been following your threads, and learning a great deal. Could you please help me to understand what is meant by an "outside reversal", and how it would differ from a regular outside bar? Thank you for your help with this.

[mw]

An outside reversal is merely an outside bar following another which, if you like, is similar to climbing a building on one flight of stairs, and descending the building on another flight of stairs, ending up in the same place as one started.

(or the reverse if selling off then coming right back up)

Probably the most important thing to consider is that outside bars, of any sort but particular the type with a tall "real body" (in between the open and close) which envelop some or all of the prior bar (again, if that was a tall bar so much the better) -- are generally rather violent things. For example so when price closes up one day (investor giddiness is then rampant), opens up a little higher (wow, we are making more money!) and sells straight off only to close lower than the previous low (oh my gosh, what happened!) - that sort of price action is rather demoralizing.

Its not unusual to see them near swing turns, even if the outside / outside reversal bar isn't the actual turning point. Anytime investor sentiment is being turned on its head is a time to pay extra attention.

So far the Dow sell off has been very directional. A 1,000 points peak to bottom and still very little recognition that the trend has changed. Bizzare.

Anyway... there's plenty of room for the Dow to bounce at present; a minor lower open tomorrow that doesn't follow through could easily set the stage for an outside reversal. I almost expect such a thing, based on the attempt today.

But even so, I do not currently expect any market bounce from these levels to last more than a few sessions. That's more gut feel than anything, its certainly not based on measurement.