Three Line Break Charts

 I'm not even going to post any regular charts this week, because I'm tired of reading about COT reports and gold and silver bottoming. A relief rally is likely, perhaps a strong close to February, but I'm so confident a new low will be made within the next month or two, that I'm gonna take some time off until then, posting only a special kind of chart.

My decision also has to do with some poker wisdom I picked up back in the day: you gotta leave the table when you're on a long streak, whether it's a bad streak or a good one. Both promote reckless decision making.  I've been "on" so far this year, but the phenomenon of "winner's curse" implies that streaks almost always have a heavy dose of luck involved. I want to step back and digest the big changes that have occurred recently, with QE3 and QE4 somehow an enormously bearish force thus far. Although I see a new low being made, there are obvious price points waiting right below (such as $1525 and $27.50) that probably will limit the downside; if not, by that point it would be dangerous to short anyway since (as I believe) the fundamentals may "snap back" at any moment and leave the retail shorts looking for their teeth on the ground with both eyes swollen shut.

Ok, the chart I wish to unveil is the Three Line Break (TLB) chart, which I have never seen anyone use before, but maybe I haven't looked around enough. I've been playing with them for a long time now, and they really seem to be great for visualizing long term trends and critical price points that may signal an end to a given trend.

Like normal charts, they of course have price on the vertical axis. But on the horizontal axis, "time" is not measured in constant units (e.g. days, weeks, months) but rather in extension/reversal events. Of course, you do have to set a certain time frame (e.g. days, weeks, months) within which the extensions or reversals can occur. So, let's say we choose weeks as our time measure. Then, if the closing price of a week is neither a reversal or an extension, there will be no additional "tick" on the horizontal axis. Time stands still, such that a whole year may be missing from a chart. On the other hand, a year in which many weekly closes extend a current rally will look stretched out on the horizontal axis.

Let's look at the monthly TLB silver chart below to illustrate how they work. First thing to remember is that these charts always reflect closing prices only. I've chosen green bars to reflect an uptrend. If the end of a month sees a new high relative to the current green bar (which reflects the closing price of an earlier month), you get a new green bar, with a new month ticked off on the horizontal axis. 

Reversals are a bit more complicated, but for good reason, as we'll see. If the present bar is the first or second green bar in a series, then price has to close a month below the lowest point of the previous bar for a reversal (red) to occur. If however, three (or more) consecutive green bars have been built, then the next bar will be red only if a month closes below the lowest point of the first of the three most recent bars. Hence the name, Three Line Break. The analogy for red bars is just what you'd expect: if three consecutive red bars have been built, that means that the bear trend won't end until price closes a month above the highest point of those three red bars. Note that however many bars are created after three in a row, only the third from last has to be surpassed.

Intuitively, three consecutive bars of one color can be understood as a confirmation of a micro-trend, such that there needs to be stronger evidence before one can confidently say it has ended. 

 Looking at the monthly silver chart below (I am re-pasting it from above):

we see that these charts are great at capturing micro-trends, macro-trends, and momentum too. We see the years 1994-1998 completely collapsed: silver was dead money. The bull market began in 2003 (unlike gold's, which really got off a year+ earlier). There have been three corrections since. The first, in 2004, was minor: only one red bar, reversed by the end of the year. The one in 2008 was of course huge, such that it took all of 2009 to reverse it (making 2009 a very short year). The third correction occurred after the April 2011 run-up, ran out of downward momentum and reversed in August 2012, right before the QE3 announcement. It extended up further in September. It has not yet been broken, though it is currently being threatened.

Note that three consecutive green bars have not yet been built, so the end to the current 2-year silver correction has by no means been confirmed, as promising as the chart looks. We would need a monthly close over $34.56 to get a third bar, which would probably signal new highs ahead. If price closes a month below $27.45 (creating a new, long red bar) that would be an extremely important sign that silver isn't going anywhere anytime soon -- at least in the up direction.

Now, the monthly gold TLB chart:

Because the post- QE1 rally has been so strong, a month would have to close below $1435 for a correction to even begin. On the other hand, given the $1826 monthly close in 2011 … 2012 isn't even on the chart. It was the first year of this bull market in which nothing important happened.

 The monthly TLB charts are thus not generally useful for short-term trading except at rare price points (many of which we approach now). The weekly charts are a little less "clean," but have better resolution. In silver, we see that a reversal of the current post-October 2012 downtrend will require a "three line break", with price needing to close a week over ~$33.75 before a green bar can be built. On the other hand, as long as we don't get a weekly close below ~$27.50, a nice inverse head and shoulders pattern will have been built.

Similar conclusions can be drawn with gold.

I'll just leave you here with some of the other charts I look at in TLB form. Check out the weekly "yields in silver" and "yields in gold" TLB charts: in both cases 2012 was a very short year. The Fed found a sustainable price level for "real" yields and just held it there.

The bullish three line break in silver yields occurred in September, while in gold (below), one occurred in June. Surprisingly, that happened right as gold was making its yearly low. Few saw what the Fed did there, though the TLB chart reveals it: despite gold's plummet, yields became convertible into fewer gold ounces than ever before. Note that both these charts are in danger of a non-confirmation reversal; I've been following them closely and I'll keep you posted.

 Finally, check out the monthly TLB chart of yields measured in oil. This month will close with a third green bar (actually yellow in this chart). The very long term pattern has been a sustained decline that loses momentum, and is subsequently punctuated with ~three green bar intervals. So it's probably a bullish sign for crude prices (or a bullish sign for bonds) that the third green bar (with dying momentum) will probably be in place at the end of this month.

ADDENDUM: In case anyone's read this far, I have two bonus TLB charts for Victor the Cleaner: the crude and gold ratio. I also included one for BRENT but for some reason prices are only available since 2006.


Anonymous said...

Thank you, this is fun. No action in WTI since 2009. Last action in Brent just before summer 2012. That's very true.


Anonymous said...

Have a rest.


"May the charts be with you".

Louis Cypher said...

Awesome work GM.
Been pondering the question of what commodity to compare gold to measure against through history. Using a mans suit is just plain silly.

Two things occurred to me ... Livestock and of course whatever the the favored currency of the day is.
Livestock because it is always needed and the advances in production have been modest (like gold mining) compared to say wheat. Currency except in times of transition in the flavor of the decade.
Livestock would seem the better of the two because we don't print cows yet.

Anonymous said...

Louis, this is why I always insist on the gold/oil ratio. That's the dollar's reserve (old) versus the Euro's reserve (future).