Monday, December 3, 2007

Data Grim for U.S. Buck

by Alex Roslin
Monday, December 3, 2007
Kitco.com
[original article]

When The Economist ran a story on “The Panic About the Dollar” on its cover last week, some contrarian traders thought, “Get ready for a dollar bounce.” The thinking is based on what can be called the front cover indicator—by the time the major media picks up on something, the trend has usually run its course.

The Commitments of Traders reports often work in much the same way. These free government reports tell us how trillions of dollars are positioned in futures and options markets in everything from gold to copper, the SP500 and U.S. dollar.

One of the things I found when looking at this data is that it really is true—uncannily and disturbingly so—that the crowd is usually wrong in the markets. So much so, in fact, that I figured out a way to trade off this data when traders hit specific statistically significant extremes in their bullishness or bearishness. The data is actually so often consistent, there’s no need to look at actual market prices. You can trade off the COTs alone.

In copper, for example, the large speculators—these are the big investment firms and hedgies—tend to be quite badly positioned at market tops and bottoms. You could almost feel these folks wincing in pain after they hit a historically extreme net long position in the Sept. 25 COTs report, just days before the copper market peaked and subsequently crashed. The large spec positioning at the time gave me three renewed bearish signals for copper, starting the week of Sept. 25. (For more details on this and my other metals setups, see the table included with my original post of this article at Kitco.com, and to see my signals in other markets, visit my free blog COTsTimer.Blogspot.com.)

What I find interesting now is that, while copper has gotten chopped to pieces, the large specs have very steadily built back up their net short position. Now, these guys have gotten to the point where they’re quite bearish by historic standards, compared to their past positioning. If this trend on their part continues, we could have a bullish signal before long. But I should caution we’re still not near that point right now. (I should also point out that my copper trading setup, while showing market-beating returns in past results, is less statistically robust than one would like to see to trade off it alone.)

So what about the greenback? Despite the talk of a bottom for the greenback—including the contrarian signal of The Economist cover—the commercial traders in U.S. dollar index futures have now reduced their net long position to a historically extreme low. The latest COTs data gives my dollar setup a renewed bearish signal. This caps a 10-week fall in the commercial net long position, which peaked in the Sept. 18 COTs report. This renewed signal is somewhat striking because it’s the first signal of any kind for this setup since the initial bearish signal that came way back in Oct. 2006. Look out below!

All my other signals remain unchanged from last week’s COTs report: bullish for silver, Canadian Gold iUnits and platinum; bearish for gold, the HUI Gold Bugs Index and USERX U.S. Gold Fund.

Good luck this week.

No comments: