chart of the day

Do not worry about a crash until it’s time

Last week I decided to address the recent onslaught of market crash stories by stating that they don’t happen in a vacuum and then demonstrating what I meant by that from a neoclassical technician’s perspective. There is a structure to large market declines for, after all, humans push the sell buttons and program the computers that do the same and we are all creatures of habit.

I used the ’87 crash as the example since recent crash talk has been directed at the similarities between then and now. But it is only one example of a debilitating decline. There are two other more recent crashes that most of us have both traded through and certainly lived through. The 2000 decline came after a ridiculous run higher where fundamentalist were starting to count eyeballs to value companies. It was, after all, the Dotcom world where eyeballs were as good as wallets.


And what about the more recent debacle where the financial industry divorced the risk of issuing a loan from the actual loan and sold us on the notion that it created a less risky investment. That went over well didn’t it?


Since penning that piece, my fellow colleague here on MarketWatch, Mr Gayed, has suggested that we should indeed worry about another 1987 style crash because, in his words, crashes occur “during deflationary pulses, prolonged internal distortions within and across asset classes, and sudden liquidity scares.”


Folks, booms and busts have been with us for ages and their root cause is a leveraged financial system where credit runs amok and those holding the debt finally realize that the debts they hold can no longer be serviced. That’s how deflationary pulses, prolonged internal distortions within and across asset classes, and sudden liquidity scares originate. So in this regard I do not disagree with Mr Gayed (although I’m not sure that these are the only three possibilities) but instead peel one more layer off the onion to see the real root cause.


So, are we in this situation again? Has credit been extended to the point that those holding the debt cannot service it? Uh, that’s a good question because a significant shift of debt has moved to the loose wallet central banks of the world - those institutions that have been entrusted with the great power, virtually no oversight, and a monopoly on the creation of money. And what’s even grander is that they are all locked in a race to devalue their currencies through quantitative easing in one form or another for a myriad of reasons.


Now I’m not smart enough to tell you how this grand party ends for the lender of last resort is now the holder of enormous amounts of paper with questionable value, but since they don’t have to balance their books, will the credit/debt cycle play out? And even if it will, how deep can they dig the hole?


Since I, nor anyone else, can answer that question yet we all seek to find returns in one form or another, what do you do. Do you stick your head in the sand out of fear? Well, if you really can’t tell when all these issues will create the dive that everyone is so worried about then maybe that’s not a bad choice.


But what if you could tell with reasonable certainty that the potential for a larger drop in share prices can be foreseen? My work says that you can. With neoclassical technical analysis we are not tied to the concepts of the past. We measure trend in an algorithmic fashion and as a result we can back test and draw conclusions based on the way the market has behaved. Unless you believe that human’s habits have really changed dramatically, it seems reasonably logical to look at the past when considering the future. The technical set-ups I showed which indicated the potential for a crash in 1987 existed in the 2000 and 2008 crashes as well. I didn’t cherry pick “a crash” but instead shared the research I have done over the years to show you the structure of how crashes unfold - and they do unfold!


So, using these other two more recent crashes, let me walk you through those charts and point out the identical setups with an added neoclassical technical tool, the retest and regenerate sequence.  Below are the weekly and daily charts of the S&P 500 (SPX). The weekly chart spans from February 2000 through February 2001 while the daily depicts July through October.
As with the 1987 crash notice that the break of multiple swing points on multiple time frames precedes a fast move in price in the direction of the break.  It’s kind of the straw that breaks the camel’s back.

 

In these charts I have added one additional construct that I have talked about in numerous articles which is a treasure trove for more active traders and investors since it provides an early warning sign of potential trend change. That concept is the retest and regenerate (RT/RG) sequence. RT/RG occurs any time a swing point is broken and price retraces to test the breakout area and potentially regenerate in the direction of the break. Since more than 2/3rds of all breakouts retrace within the first six bars of the breakout, retests are commonplace and their success or failure is of keen interest because, if a RT/RG succeeds it implies that the breakout was real and price will continue in the direction of the breakout. A failure has, of course, the opposite implication. The test can trade as far back as the far side of the swing point bar that was broken.


In the case of all these charts, the shaded area denotes the RT/RG zone. When an instrument has been on an extended run (up or down), each successive break of a swing point in the direction of the trend, and the subsequent RT/RG sequence becomes of utmost importance since a failure to regenerate is a precursor to a potential trend change.  We are currently doing RT/RG sequences on the strongest two indexes now - the NASDAQ Composite and the NDX 100. It’s the first clue of whether we should raise our eyebrows are not. It also is the first trip back on the strong indexes after more than six bars have passed since the breakout. The probabilities that it will be bought are reasonably high and we saw some of that going on Friday.


Returning my attention to the included charts, as you can see, these retest and regenerate failures provide the first warning sign. Since they occur frequently they also fail a good percentage of the time so you have to realize that you cannot jump all-in or all-out on such a breakdown but you can become a bit more careful if you care to just in case. Whether that means tightening up stops, hedging, selling off some weaker assets or doing nothing , that’s a matter of preference and risk tolerance.
What is not a matter of preference though, is a break on multiple swing points and time frames. In those situations a trader and an investor is required to become more cautious and protect themselves - immediately! In 2000, had you not done so at the 1410 area, you would have suffered through another 45.5% drop in price over the next two years.

I’m sure by now you realize that 2007-8 was the same situation. The set-ups are the same. Here’s the weekly chart with the 1400 area again being the key breakdown point and with multiple clustered swing points apparent just below.

Although everyone refers to the crash as the 2008 crash it really began in 2007 as shown here in the daily chart that spans from May through July of that year.

Once more we see the early precursor bullish RT/RG failure and then a break of multiple swing points around the 1485 price area which, when looking at the weekly chart above, led to the first big scary spike down in early August.


After that, prices proceeded to climb all the way back to the highs one more and the next drop was a clear bullish RT/RG failure on the weekly chart which eventually led to the second scary and sustained drop that took out multiple swing points on multiple time frames. Here’s how the daily chart looked at that time.

Again, the set-ups are the same; bullish RT/RG failures and bearish RT/RG successes usher in lower price. And just like in 1987 and 2000, if you did not move out of the way or flip to the other side of the market when these telegraphed breakdowns began to occur, you were slaughtered along with all the other sheep as shown below.

Now Mr. Gayed admits that it is virtually impossible to predict these crashes with his tools and methods, thus he suggests that you worry about the possibility. My take is, as it was in the earlier article, worry about it when it’s time. There are always issues in the market. One can always find something to point to and worry about. The market is comprised of so many issues and relationships that you can almost always find something to worry about if you so choose.


My approach is simple. If the technical characteristics begin to form, then worry.  Personally, I don’t care if the credit/debt cycle manifests itself in such a way that a deflationary pulse is what starts the price snowball down the hill. It is unimportant to me if a sudden liquidity scare is the culprit. It may end up being some other root cause. What we do know is that the characteristics will begin to form on the charts because, whether you like it or not, the charts are a reflection of all known knowledge since they are ultimately based on what the cumulative body of traders and investors are willing to pay. Thus, what I care about is being able to identify the fact that the probability that prices will cascade lower is becoming significant and then take measures to protect against that increased probability and with neoclassical technical analysis, you can see it coming.


It’s not as if a crash sneaks up on you in the dead of the night. As I said early this week, and I say again now, the current market currently does not have this structure - at least not yet. Sure it can develop but the operative word is “develop”.  It takes time. It’s like running out and buy water bottles and batteries because hurricane season is approaching - before one even forms off the coast of Africa. You have to separate the possible from the probable with respect to the near future. It’s not probable right now. When it is I’ll change my tune.

This article was originally published on MarketWatch on Aug. 19, 2013, 12:20 p.m. EDT