This essay is based on the Premium Update posted September 5th, 2009

It was a blockbuster summer for the bulls on Wall Street but September is historically the market's worst-performing month and already as the month has just began we've begun to see some sharp pullbacks. Interestingly enough, on the flip side of the coin, September is historically the best month for gold.

I mentioned buying precious metals mining stocks two weeks ago - if you did, you would have made about 14% profit so far if you diversified in many stocks (HUI Index), and about 19% if you put your money in the top 3 stocks that my leverage calculator suggests as a speculative proxy for gold.

The price of gold has risen in 16 of the 20 Septembers since 1989. And as this September began we have already seen some tangible proof as the yellow metal makes new forays toward quadruple digits.

Source: U.S. Global research

At the moment gold is on course for its biggest weekly gain since late April.

While September is a good month for gold, it is historically a great month for gold stocks as measured by the NYSE Arca Gold Miners Index. After the typically weak summer months the gold miners start to perk. Since 1993, when it was created, the GDM has been up 11 times in September and down just five times.

September is also historically a miserable month for the U.S. Dollar, a bullish sign for gold since gold trades at an inverse relationship to the dollar (More on that later.) Looking back 39 Septembers going back to 1970, the dollar has seen negative performance 26 times, more than any other month of the year.

So what's behind this predictable September pattern, as far as gold goes?

Several gold-demand drivers converge all at the same time this month.

• The post-monsoon wedding season begins in India..

• The Indian festival season begins.

• American jewelers begin restocking in advance of Christmas.

• Ramadan ends in late September in the Muslim world with a period of celebration and gift-giving.

• And, in China, the week-long National Day celebration starts October 1. Already in China gold jewelry demand increased 6% in the second quarter as we reported in a recent Premium Update.

While just about everyone in the world is getting ready to celebrate, the price of gold (charts courtesy of stockcharts.com) is getting ready to perform its own celebration dance.

In the Market Alert that I sent out on Thursday I pointed out that we have just seen a significant breakout from gold triangle pattern to which I had referred in the past. The move took place on a very strong volume. This is exactly the type of confirmation I like to see in a breakout.

The RSI Indicator, proven to be a very valuable tool in timing the gold market local tops, suggests that gold may need to take a breather before moving higher. Should that be the case, the most probable scenario is a test of the upper border of the triangle pattern.

Still, in the short term prices may rise a little higher before correcting.

Analysis of the short-term chart suggests that although prices have risen high and fast this may not be the end of this rally. First of all, the volume is not low. It was lower on Friday than during the previous two days, but it is still considerably higher than the average during the past few months. Low volume would indicate that the buying power has dried up and price is ready to plunge. We do not see this yet.

Also, please take a closer look at what happened at the end of May. The RSI was very close to the 70 level in the 4th week of the month, but the price pulled back only for two days, and then rallied further. While I can't say for sure that this will happen here, I can say that there are no clear signs of a top yet.

Moreover, taking into account the 1.618 ratio that is also very useful in predicting the range of future price moves, we might expect the GLD ETF to reach levels marked with a red rectangle. This would correspond to gold breaking above $1000. Should this move materialize, I would expect it to be volatile.

Gold and gold stocks move together most of the time (what is confirmed by high correlation coefficient values between gold and HUI in the correlations table), so we can say something about gold by analyzing the performance of gold stocks - here - via the Gold Miners Bullish Percent Index (a market breadth/momentum indicator that is calculated by dividing two numbers: the amount of gold stocks on the buy signal (according to the point and figure chart) and the amount of all the gold stocks in the sector).

The GMBPI itself is currently trading at the overbought levels that in the past sometimes meant that a top is in. However, that is only part of the story.

Please remember that if during the second half of May 2009 an investor hastily acted solely on the popular use of the RSI that says to sell once it gets above the 70 level, he would have missed a $50 move in the HUI Index.

It's usually a good idea to put every indicator/tool into the proper perspective before acting on it. Keep in mind that the people who design an indicator want it to work in many markets and usually don't fine-tune it for a particular market. Investors need to make the appropriate adjustments themselves.

Here, when you consider the local tops that materialized when both the RSI and William's %R indicators were in the overbought territory, it becomes clear that these tops did not form immediately after the overbought levels were reached. Conversely, the HUI's value usually moved higher for several days/weeks before topping. I have marked with vertical, dashed lines each time gold stocks took their time before reacting to the "overbought-sell-now signal".

So, even though the probability of a correction increases it does not mean that it will take place very soon.

Summing up, your approach should depend on your investment perspective and risk preferences. The trend is up for the PMs, but there are a few signs that some kind of breather is likely. This does not need to take place instantly, so the question is whether or not to hold your positions in this market.

The fundamentals have not changed, so I don't think that selling one's long-term PM investments is a good idea right now, however the short-term speculative capital is another matter. If you are risk-averse, I would suggest exiting at least a part of your speculative positions right now, and re-entering during a consolidation. Those of you, who accept high risk in order to reap biggest gains, may want to wait for PMs to move a little higher (to levels mentioned above) before closing your speculative positions.

To make sure that you get immediate access to my thoughts on the market, including information not available publicly, I urge you to sign up for my free e-mail list. Sign up today and you'll also get free, 7-day access to the Premium Sections on my website, including valuable tools and charts dedicated to serious PM Investors and Speculators. It's free and you may unsubscribe at any time.