Wednesday, May 6, 2009

Gold Gets Its Groove Back For Now

by Alex Roslin

Kitco.com

As stock markets fly up from their March low, gold has slowly wilted. Even a decline in the greenback hasn’t returned bullion’s shine. Gold seems to have met solid resistance around $1,000—the level it hit in Feb. 2008 and again this past February before taking serious tumbles.

Could this be what technical analysts call a double top—a bearish formation that usually leads to a major selloff? Or are the gold bugs right to say gold is about to explode because of the oceans of central-bank liquidity being mainlined into the financial system?

The questions are especially important as gold has recently tended to move in the opposite direction to the stock market. So if the equities rally is for real, will this spell trouble for gold bulls?

We can find some interesting answers hiding in some little-known derivatives data that comes out once a week in the U.S. government’s Commitments of Traders reports.

I know, I know—talk to most people about derivatives and you’ll see eyes glazing over and hear the sound of snoring. Well, get another coffee because this data tells us where the big players in 100-plus markets are parking their money.

And in the gold market, things seem to be looking up—for now. My trading setup based on the COT data gave a bullish signal for gold bullion for the open of Monday, April 27. It had been in cash before that for 11 weeks.

The signal is based on a couple of key developments in the data. Firstly, the so-called large speculators, who tend to be wrongly positioned at key market junctures, have recently gotten quite bearish on the prospects of the yellow metal.

In fact, starting in early March, the large specs sharply reduced their total open interest (long plus short positions). This caused them to fall significantly below a long-term moving average I’ve developed to track where they stand compared to recent data.

Meanwhile, another positive development has occurred in the large spec net position. It has remained in a highly pessimistic posture since last August. That, of course, is also bullish for gold bugs. We want to get long when the dumb money hits extremes of gloom.

And last August, the wrong-way large specs hit a historic peak of bearishness that they have yet to really reverse. In fact, the week of Aug. 19, 2008, saw them get more bearish in their futures and options net position as a portion of the total open interest than they’ve ever been since the beginning of the data, relative to their moving average.

Since then, it should be said, the large specs have gotten somewhat less down on gold. But they haven’t gotten anywhere nearly frothy enough by historic standards to jeopardize a gold rally. That was even true during February’s spike past $1,000.

My trading setup based on the COT data works by combining two signals and taking positions in bullion when they agree. The first signal is fades the large spec total open interest in gold; the second fades the large spec net position.

I’ve found through lots of backtesting (including Monte Carlo tests and detrended price data for you stats enthusiasts) that trading with two signals based on this data tends to be far more reliable than trading off just one signal (my initial method when I first started researching this data a few years ago).

It’s not clear how long the current bullish gold signal will last. Each Friday afternoon’s data brings the possibility of a new signal. One warning sign has already shown up for a few weeks out. The large specs total open interest shot up suddenly to a bullish extreme last Friday. However, this signal works with a seven-week trade delay, so it doesn’t affect anything for a little while yet. Just a warning sign that any coming rally might get overbought real fast.

I should also point out that my signals are no guarantee of a winning trade. No trading strategy I’ve ever heard of wins 100 percent of the time; mine certainly hasn’t, even in the hypothetical realm of backtesting. I use stops and appropriate position sizes to help control my risk.

Visit my free blog COTsTimer.Blogspot.com for my weekly updates on this and other markets, to see how on how my system works, to learn more about the importance of risk control and to download a spreadsheet to crunch the data yourself.

***

Disclaimer: This report isn’t meant as financial advice or a recommendation to buy or sell any security. Please do your own homework before trading. My system involves substantial risk, has experienced large drawdowns in some past trades and requires the use of additional risk-control techniques. Past results are no guarantee of future profits. I’m not a certified financial advisor. While I consider my information to be reliable and accurate, I make no guarantees. Visit my blog for more important disclaimer information.

No comments: