Tuesday, February 28, 2012

Gold's Big Fractal Fork In The Road Revisited

Back in 2010, I was following the fractal analysis of David Nichols on gold. He was a gold bull, but he was predicting the end of the gold bull market to occur around February of 2011. I was a gold bull too, so his prediction filled me with great consternation. I had been going over the past predictions of Nichols and knew that he was stunningly accurate most of the time, especially in the mid-range time frame of a few months or so. It made me wonder just what he was smoking. I studied the reasons why he was predicting the end of the gold bull in February - and found them to be very convincing.

I'm going to revisit what he was saying back then, and go over some additional fractal considerations I had found that gave me hope for the gold bull surviving past February a year ago. Allow me to excerpt two articles I wrote in 2010 "The 64 Month Bull Market Fractal" and "Gold Is At A Big Fractal Fork In The Road".

From the first article on the Nichols fractals:Link
I don't know how familiar you may be with the emerging science of fractal market analysis, but there is an element of it that has a direct bearing on gold right now. To briefly overview fractals, they have found that stocks and indexes have a strong tendency to move in repeating chart patterns at various scales (self-similar, they call it). So a stock may consistently produce say a 2 year pattern which is also evident at a 2 month time frame. Sounds silly, I know. The theory is that there are well known fractal growth patterns in nature, crystals growing under a microscope and about any basic growth in nature; and these are abundant with self-similar, geometric, repeating patterns. They've known about these nature patterns for many decades, but only recently has anyone thought that financial markets may grow by these fractal patterns too. When they investigated, they found that the unchanging human nature did indeed infuse fractals into the trading charts. David Nichols, a pioneer in this investigation, finds that there is a 64 month parabolic fractal signature that seems to show up at about every major bull market. The duration varies a couple months or so, but the pattern is a "sprouting" of a parabola, a bullish change in previously sleepy trading, followed by parabolic growth into a violent top about 64 months later. As an example, he points to Toll Brothers as a proxy for the housing bubble (click to enlarge charts):



The months are marked 1 through 65 for the parabolic progression. The Egyptian stock market did the same thing. I won't flood this piece with the other charts, but the same thing shows up in the Dow of the 1920s, the Nasdaq of the 1990s, and other bull markets. "

To see other examples of this 64 month parabolic fractal, just click the link above to the rest of the article. Nichols equated the end of this parabolic fractal to the end of a bull market. But I have found many examples where the fractal did not mean the bull's end - see article. Oil is one such case. Remember oil's recent parabolic run? It was from a sprout point in early 2003 going to the top in 2008, and it ran 63 months ! Does that mean we'll never see $130 a bbl again anytime soon? Don't bet on it. Bull markets can do the parabola fractal, cool down, then go into other forms. I show several examples of this in the article.

In September of 2010 when I wrote the above article, it certainly appeared that gold was going to be yet another case of 64 months, then a big cool down. Nichols portrayed the fractal thusly:



If you count out the months from a Sept. '05 parabola origin, you arrive at February, 2011, give or take a month or so. But as I pointed out, that end point for the gold bull creates a lot of misfits in the big picture. We are nowhere near the mania end of a parabola. Investor ownership of gold is still near historic lows, around 1% of assets, compared to historical norms of 5% or more. And as I complained in 2010:

"Nichols reckons the change in behavior "sprout" point as September 2005, although gold was already in an uptrend by then, but a very weak one. Since this chart was done, we have pretty much been following the fractal.The end of this gold parabola is early 2011. This brings up some difficult questions. If the gold bull market is to end in 5 months, does that mean the entire commodities bull also dies in 5 months? It's hard to imagine a commodities bull without gold being a part of it. Do all the world's debt and currency problems go away in the next 5 months?...

In a September, 2007 statement, he [Nichols] says, "The most important investment theme for the next 10 years will continue to be the frenzy for tangible, hard assets ... and the best market to take advantage of this monumental trend is gold."

So here is where a fork in the fractal road is developing. Those of us who see a continuation of the currency/demand induced commodity bull market for years may have trouble digesting the 64 month bull market fractal that Nichols has recently posited as beginning with the parabolic sprout in September 2005 and abruptly ending in February 2011. It's enough to give a fiat money hater indigestion. In my article linked above on the 64 month bull fractal, I point out the amazing prevalence of this fractal - it's everywhere. And gold is showing all the signs of following this pattern. So do the world's debt and currency problems all go away over the next two months? Does the commodity bull cycle suddenly end now? Has Nichols changed his mind about the investment theme for the next 10 years? In 2008, he stated "this bull market in gold should last for many more years." His 64 month gold fractal seems to contradict this...

So what right does this fractal stuff have to upset the applecart of sound fundamental considerations? Maybe we should just forget all about this fractal hocus pocus. Well, not so fast. Nichols seems to have only recently stumbled upon the 64 month thing. But just maybe there are other scales of this same fractal out there that we should be thinking about.

First, you have to consider the basic structure of this fractal. It is composed of two parabolas separated by a distinct downtrend phase about midway. For examples, look at the currency parabola of 1920s Germany and the stock market of Denmark in the 90s: (click on charts to enlarge)




The shapes of the components vary, but the structure keeps recurring with the downtrend in the middle being a year or less with the overall time within a couple months or so of the 64 months. This is by far the most common scale of this fractal. But it does seem to occur in other sizes. There is a 3 year version. As examples of this, look at Homestake Mining, the premier gold miner of the '30s, and the Brazilian inflation of the early '90s:



These fractals do the same thing in different scales. That's what fractals do. They are all different with the only common denominator being that all the main components are sized proportional to the overall size of the fractal. The 3 year mid-course downtrends are small, less than a year, and barely noticeable; but they're there and much more clearly seen in the amplified versions of this fractal.

Could it be that there is a much larger scale of this same fractal that gold may actually be following in lieu of the common 64 month size? Does such a thing exist? It very well may. Look at these examples: the Thailand stock market of the late '80s and early '90s



The large size seems to cluster around about a 9 year length with everything scaled up including the variance on how long it runs and the duration of the downtrend in the middle, which runs around 2 to 3 years. The next example is the stock market of Turkey



This 9 year iteration has the mid-course downtrend run a whopping 3 1/2 years. This is also an example of a parabolic rise not meaning an end to a bull market as is commonly thought. The Turkish market could hardly be considered busted when the parabola was over. The next king-size example is the Australian dollar



Given that the downtrend size in the middle ran a little over 2 years, you get the impression it wanted to run to more like the 9 year length before being interrupted by the end of the world in late '08.


... the Swiss stock market of the '90s



There is one other pertinent example of the large version of this fractal, and that was gold in the '70s. It featured a two year downtrend in the middle in 1975 and 1976. The length of this downtrend is about the most reliable predictor of which scale fractal is being carried out. The distressed buy and holders of gold during those two years would like to have known this basic fact back then. Well, we here in 2009ville know. This time around, the mid course downtrend gauge has been completed as of a little more than a year ago, and it was nearly two years, suggesting that it is indeed the large scale fractal we are in, replicating the previous gold bull"




It's been a year now since the fork in the road was encountered, and it appears the larger scale fractal is what we are in. We are tracking the second of the twin parabolas. This was the projection I was making back in 2010. Here is a closer look at what gold has done:


One thing that strikes me about this fractal thing is that it projects exactly the same thing for the future of gold that Jim Sinclair does, as my my article on his model details. Jim Sinclair does not base his highly accurate projections on fractals, so these are two very independent means of analysis pointing to the same thing for gold - the high road.