Sunday, 20 November 2011


This is the Year of the Cat.  

No, I'm not talking about the mysterious cat in the Vietnamese zodiac.  And I'm not talking about the excruciating 1970s album which gave Al Stewart his one and (thankfully) only hit.  I'm talking about a strange creature which stalks the corridors of science simultaneously dead and alive, a feline paradox known in the world of physics as Shrodinger's Cat.  


As we stand in late 2011, the world's moribund, debt-addled economy seems to many observers like a cat on its proverbial ninth life.  

Hemmed in by crippling debts, ailing banks, consumers in lockdown and a failed currency system dragging the world's biggest economic bloc down into depression, the cat finally seems to be facing its nemesis.  In fact it looks as though The Perfect Storm I envisioned in these pages 15 months ago is at last gathering full strength.

But rather like Shrodinger's famous quantum moggie - which, until it's observed, exists in a state of trembling suspension between life and death - what's fascinating about recent market behaviour is that it offers up two simultaneously contradictory prospects.  Because a balanced reading of the technical picture suggests that what we may be looking at is not the beginning of a descent, but the end of one.  

Now, my views on the economic outlook are unambiguous and have been laid out here over two years for all to gasp at in amazement when I'm right, or point and laugh at when I'm wrong.  

But in making investment decisions, I try to suspend my opinions and make judgements based on objective, measurable, statistical fact.   

And what statistical studies of 100 years of market history appear to suggest is that we're right at the fulcrum point between the best of times and the worst of times.  We may well be at the start of a new and frightening leg down; yet the possibility also exists that, in October 2011, we saw a major low in sentiment and stock prices and that those lows will never be seen again.

If that's the case, we are looking at the most incredible long term risk / reward opportunity for investors since the  bear market trough in 2009. Should that reality not soon materialize, however, the opposite risk/reward skew, to the downside, is equally dramatic and compelling.  

So to every dyed-in-the-wool permabear and sunny-side-up ultrabullI suggest it might pay to suspend your growling & snorting for the few short minutes it takes to read this post. 

Because only one of you is going to be smiling in 2012. The good news is, with distant targets in sight whichever direction this particular bazooka is fired, a smart investor only need hang on to the rocket.  It's therefore possible that an opportunity to make life-changing returns with very low relative risk lies just ahead. 

Let's look at the evidence.


To show you my reasoning, I'm going to need to lift the lid a little on my box of trading tricks.  

Most of my buy and sell decisions are filtered through the lens of volatility - if I can see a market is about to make a sizeable move, I'm interested.  If not, I stand aside and let it chop.  

What's interesting right now is that several indicies, when looked at on long term monthly charts, are setting up to make very sizeable moves.  Which means 2012 is likely to see huge gains - or losses - one way or the other.

Exhibit 1: Germany's stock index, the $DAX

$DAX, MONTHLY, 1994-2011

& Bollinger BandWidth volatility indicator

I use a Bollinger BandWidth indicator, itself surrounded by Bollinger Bands, as a somewhat crude but highly effective volatility gauge.  When the indicator falls below its own lower band the market's volatility is dropping substantially below an almost two-year mean.  And because volatility is cyclical this leads, reliably, to a long period of higher volatility ahead.

The green circles highlight periods in the past 16 years when monthly volatility is as low or lower than it is today. They were truly excellent times to take positions in the DAX, whether long or short.

In addition, the red circles in 1995 illustrate how a contraction of the Bollinger Bands can often indicate an especially powerful move is building.  This phenomenon is occuring right now and implies a sustained and powerful trend ahead.

S&P500, Monthly, 2004-2011 
The DAX is one of a number of stock indexes which have a similar volatility profile. The S&P500, you may be interested to note, is another.

Exhibit 2: Mismatch between 
current and implied volatility

Jason Geopfert of Sentimentrader (whose brilliant work I'll reference again later) tested for periods when, as recently, the S&P500 was mired in a range and chopping back and forth, while the activity of options traders implied considerably higher volatility to come.

He looked for a very low ADX directional movement indicator (11 or lower) and a simultaneously elevated VIX volatility index (22 or higher).  There were few instances looking back 25 years, but in each case sustained, major moves began almost immediately.

June 1987

May 1988

March 1999

February 2001

December 2001

In some cases, the market dodged back and forth a few times before deciding on its ultimate direction (and in 1987 you got a huge swing both ways) but any investor willing to duck and weave, placing trades to ride both the upside and the downside would, in the end, have been rewarded mightily for their patience.


Waterfall declines such as we saw in summer 2011 only come around once in a blue moon.  When they do, trading geeks like me get very excited because they throw up all kinds of rare historical parallels, and these can help enormously in predicting what markets are likely to do next.

Why?  Well, we're not talking about your bog-standard sell-off here. These moments capture a market in true extremis, when it's not only moms and pops from Boise Idaho but major Wall St firms and ultra-slick hedge funds who are offloading whatever stock they can dump at whatever price they can get.  

Fear of ruin is an immensely powerful motivator and in the face of it people do entirely predictable things.  After all, panic tastes the same in November 2011 as it did in September 2008, or in October 1987, or October 1929.  

And because investors react like herd animals in such circumstances, and since there is always a point somewhere at which panic subsides and vultures swoop to buy, the aftermath of a meltdown twenty, forty or even eighty years ago can look very much like the aftermath of a meltdown today.  

Exhibit 3: Waterfall declines are highly predictive - they mark either the end of a bear market, or the beginning 

Dow Jones Industrials, Weekly, as posted August 24th.
The week highlighted by the green dot was spent entirely below its
lower Bollinger Band - that's RARE panic.

During the August meltdown, I took a look at a rare chart pattern I'd discovered which strongly suggested a low was imminent.

You can check the details by clicking the link, but essentially this quirk identified a period in which selling pressure hit an unsustainable extreme.  The desire to liquidate was so intense that prices dropped more than two standard deviations below their six-month mean and stayed there for an entire week.  This is surprisingly rare. Looking back over eighty years of data, these moments of unalloyed panic were seen just eight times.  In each instance, stocks rebounded substantially soon after.

1929 & 1930: Stocks rebound from a waterfall decline.
First time they rose all the way to the upper Bollinger Band.
Second time... they never made it.

This particular analysis turned out to be correct, but another aspect of the data may now be worth noting.  

  • Of the eight instances, two - 1949  and 1987 - marked the very end of a decline, setting lows which would never be breached again. One low, 1960, was penetrated very briefly two years later, before another huge bull market began.

  • The five other instances marked the beginning of a longer-term decline, where prices recovered initially, then fell substantially, over the next several years.

  • All but one instance saw prices rebound all the way to their upper Bollinger Band, two standard deviations above the mean.  At that point the outcome was binary - major bull market or major bear market.

  • The one failure to reach the upper band was in 1930, when the Dow struggled up to its 20-week moving average then rolled over into the depths of the depression.  As I write, the Dow is struggling to hold above its 20-week moving average.

Exhibit 4: Powerful October rally suggests a binary outcome in 2012

Having examined the summer meltdown and its implications, let's turn our analytical gaze onto the October melt-up.

Presenters on CNBC almost wet their pants with excitement, headline writers had a field day and even permabears had to admit it was a stunner.  At almost 17% trough-to-peak it was the best October for the S&P500 in 37 years, and the best for the Dow Transports index in 72 years.  

Dow stocks surge in October 1939, then...
But before we get too excited ourselves and start extrapolating stock prices up towards infinity, let's remember what happened next in those two cases: while the October 1974 bottom marked a major low which would never be seen again, October 1939 proved to be the absolute peak of a rally - after which stocks dropped 40% and the economy tumbled back down into depression.  

All this reflects a phenomenon technicians have long understood: huge surges in price, whether on a grand scale like we've just seen or on a miniature intraday timeframe, are usually moves of either initiation or exhaustion and mean the rally is either the very beginning of something big, or the very end of it.  Which leads me on to...

Exhibit 5:  Blast-off from low = 
binary outcome in 2012

Over the past several weeks, Sentimentrader has examined this kind of blast-off rally from various angles, in several studies covering a century of stock market history.  

One I found particularly noteworthy.  He checked back to 1928 for previous instances of

  1. a 52-week low, then 
  2. a 9%+ rally in the space of 7 days, in which
  3. none of those days lost more than 1% 

These were the same conditions - essentially a buying stampede - that we saw in the week after the 2011 October low. Amazingly, just five instances crop up in eighty years.  They were: Oct. 1974, Aug. 1982, Aug. 1984, April 2001 and March 2009.

Well, we all know what a great buying opportunity 2009 was.  I've marked up the others for clarity:




Exhibits 6, 7, 8...

There are other objective measures I could point to which suggest either an incredibly positive or negative year ahead, but I'd rather not overplay my hand. Oh alright then, here's a flavour. 

Check out one indicator of market breadth with a great track record which is screamingly bullish.  On the other hand, here's one reliable gauge of market sentiment which suggests there's still too much speculative fervour to forge a lasting market low.  

Election cycle: 2012 = Year 4
Followers of the Presidential election cycle and market seasonality insist a rally is likely at least into next spring (and one widely-respected hedge fund manager, ace market timer and long-time bear has turned bullish), while devotees of chart patterns and pure technical analysis point to a major head-and-shoulders formation, Dow Theory and market analogs to suggest the bear market is only just beginning.



Now look, we're not just talking any old 'down or up'. Large, sustained market moves are a trend trader's wet dream whichever direction they take and fortunes can be made riding them (John Paulson being just the most recent famous example).  

Various types of sophisticated instrument exist (eg. options, warrants, CFDs) to help experienced players take a substantial punt without incurring too much risk.  But if you're not familiar with these and don't want to miss out on what may be a tremendous opportunity, I'd suggest sticking with a policy of K.I.S.S.

While of course no guarantees are possible, here is one strategy which will capture a large move, get you out in stages as it reaches exhaustion and stop you taking any unnecessary risks. 

Set up a daily chart* of the FTSE100, S&P500 or the Dax, and overlay two sets of Bollinger Bands - one with with a setting of 200, 3 (a 200-day simple moving average with bands 3-standard deviations above and below it) and the other with a setting of 200, 2.1.  

You're looking to capture a move from the average all the way out to an extreme.

  • On a day when price crosses and closes above the 200-day moving average (where prior day closed below), buy an ETF which tracks your chosen index higher**.  
  • Realize a small portion (eg. 10-15%) of your gains each time price pierces the upper Bollinger Band at 2.1 standard deviations.  Otherwise,
  • Sell your entire position if price moves back below the low of the day you bought  

This way your risk is limited to the distance between the point at which you bought and the low of that same day, while the potential reward is far, far greater - degrees of magnitude greater in fact.

When prices start moving higher, be sure to raise your stop (the level at which you'll sell if price fails to hit your targets) every once in a while.  This is vital to control risk and prevent you losing a large portion of your gains.  

But don't do it arbitrarily.  Look for a level to which prices fell, found support, then reversed back up and closed higher than the prior high.  Position your stop order just underneath the low of that move.  Be careful not to carry out this manoevre too often - wait for decent pullbacks (which don't occur more than once every couple of months).  

Either the trend will continue for so long that you cash out all your profits, or eventually the market move will exhaust itself and take out your stop.  Whichever way, you win.

Here's an example, from those I highlighted in Exhibit 2, which shows how you would have traded the 1988 bull market using this method.

Dow Industrials 1988: a simple trading strategy

So that's the dream rally - but how would you trade the strategy if we end up in a bear market?

  • On a day when price crosses and closes below the 200-day moving average (where prior day closed above), buy an ETF which rises in value as it tracks the index lower.  
  • Realize a portion (eg. 15%) of your gains each time price pierces its lower Bollinger Band at 3 standard deviations.  Otherwise,

  • Sell your entire position if the price moves back above the high of the day you bought.

It's the exact reverse of the bull market strategy - except on the way down, you make your gains three times as fast!  Here's how it would have worked in 2002:

S&P500 2002: a market in freefall

Also keep in mind that, as in the 1987 example from Exhibit 2, a trend can appear to set off in one direction and then spectacularly reverse.  As long as you are alert and prepared to follow whatever signals the market gives you, the percentage moves either way should make continued vigilance extremely worthwhile.


The simplicity of the entry method described makes understanding the strategy easy as pie, but the disadvantage is in the number of whipsaws you may have to endure as the index crosses back and forth in search of a trend.  A trader can easily become discouraged after three or four consecutive losses and abandon the strategy just before prices begin a huge move.  

You can introduce one further condition which will reduce the number of false signals without detracting meaningfully from profitability.  It requires you to enter the next day after a 200-day MA cross-over, provided there is strong follow-through from the previous day's move.  

Here's how it would work: instead of entering at the close of a day which crosses and closes above the 200-day MA, you would wait to observe the following day's action.  If the index closes above the high of that cross-over day, you have a signal to enter a long trade.  

Conversely, instead of entering at the close of a day which crosses below the 200DMA, you'd wait to see if there is follow-through the day after.  If the index closes below the low of the cross-over day, you have an entry signal to go short.  Occasionally, it may take two or three days for the high or low of the crossover day to be breached - that's fine too.  

It is surprising just how many failed signals can be eliminated using this simple technique.  But in any case, remember: when the index is at a central pivot like the 200-day moving average, knowing a major long-term move is in the cards skews the risk/reward massively in your favour; over the long course of the trend, small losses incurred at an early stage will become insignificant - as long as you don't become discouraged, don't become dogmatic in an opinion about which way the market 'should' go, and are prepared to ride whatever trend develops for as long as you possibly can.  

* If you don't currently have a charting package, offers one which is free, simple to use and ideal for these purposes.

** ETFs which track the major indicies are comprehensively listed here.  For ETFs which reward you when those indexes fall, check those in the list with a down-pointing arrow.



It seems 2012 is lining up to be a banner year.  Certainly, if this election countdown is anything like the last investors are going to be presented with a stunning opportunity, but they'll have their work cut out to take advantage of it.

With that in mind I hope this post has thrown you a little red meat-for-thought.  Because in 2012 - whatever your bias and whatever the outcome - neither you nor I will have any excuse now not to make a profit!