Gold Correction Underway... What Next?

This post was mostly written last night. I don't think the price action of today changes much and, in particular, I don't think it triggers a reentry according to the criteria at the end of the post.

That 20% run from roughly $1160 to $1390 was pretty sweet. We've had quite a party in gold since the end of July, but it appears the party is over. I don't think the bull market is over (because the fundamentals still seem strong). Here's some fundamentals that come to mind that seem bullish to me:

• The Indians have been importing with lavish premiums lately.
• Central banks are buying (mainly China and Russia) with others openly talking about joining in (google the news for Korea and gold).
• European central banks are no longer selling gold.
• The dollar is soft and there is a currency devaluation contest underway.
• The Gold Cartel which has been suppressing the price of gold is still out there suppressing away leaving gold under pressure to jump as soon as they run out of physical gold to sell. 
• Interest rates are near zero so you don't lose any significant interest holding gold rather than bonds.
• Everyone is talking about Gold being in a bubble.

So, I think its just a correction. At this point, given that I don't know when an overbought run is going to end, my policy is to wait until its clearly over (that was Tuesday morning at around 8:30AM) and close as fast as possible leveraged holdings (options and futures) and sell, on the first day, as much of my Juniors which are not core holdings as I can without personally dragging down the stock prices much. My remaining core holdings at this point are Gold Resource Corp[GORO], Rio Alta Mining [RIO.V] and Dynasty Metals And Mining [DMM.TO]). My policy is that the first day of a correction is the best day of a correction to sell. I don't want to get panicked out near the bottom.

So, what this post is about is to look really hard at the Gold charts and try to figure out a set of criteria to use for determining when this correction is bottoming. I'm going to identify similar situations and see if they can provide a guide to the criteria to consider.

Here's a rather big 5-year gold chart. The recent run-up appears as a sustained period where the RSI is overbought followed by a quick drop in the RSI to around 60 (at least). There are four (maybe five) previous similar appearences. One takeaway here is: This kind of 2-month run up in Gold is really not that unusual. It happens about once a year. And the correction that follows is also something we go thru it about once a year. If you aren't a trader you could just wait it out. The fact this happens once a year really brings into question the idea that gold is in a bubble and that now the bull market is over.If there is a blow off top to the gold bull market is going to have to be something a lot bigger than this gold run. 

The four previous big gold runup ended in: Dec-05, May-06, Nov-07 and Dec 09. Perhaps there's something comparable ending in Feb-08. So let's look at each of them in detail.

The Dec-05 correction was a minor one with gold falling around 10% (from 540$ to 490$) and in pretty much in a straight line with the bottom 6 days after the top. It:

• Did cross the 20-day moving average,
• Did not touch or cross the 50 day moving average and
• Did not touch or cross the lower bollinger band. 

The following policies would represent good re-entry points:

• First up day after the RSI  crosses below 50.
• First nice white candle.

 

The May-06 correction was a major correction with gold falling around 25% (from 726$ to 542$) pretty much in a jagged straight line hitting the bottom after 23 days (a month). It:

• Did cross the 20-day moving average,
• Did crossed the 50 day moving average (with one minor next-day recrossing) and
• Did touch or cross the lower bollinger band multiple times intraday and closed outside of the lower bollinger band for the final two days. 
• Did cross the RSI 50 (with one minor next-day recrossing).
• Did cross the RSI 30 - folks must have been sweating bullets at that point. 
• There was a period of four consecutive without a lower-low far before the bottom.

The following policy would represent good re-entry points:

• First up day after the RSI crosses below 70.
• NOTE: A fifty-day up crossing would result in being quickly stopped out for a minor loss.

The Nov-07 correction was a minor one ith gold falling around 10% (from 850$ to 770$) pretty much in a jagged path hitting the bottom after 8 days. It:

• Did cross the 20-day moving average,
• Did not cross or touch the 50 day moving average on its descent to the bottom, 
• Did not touch or cross the lower bollinger band,
• Did cross the RSI 50.
• Did not cross or touch the RSI 30. 
• There was a period of two consecutive without a lower-low before the bottom.

The following policy would represent good re-entry points:

• First up day after the RSI crosses below 50.

The Feb-08 correction may not really be comparable to the current correction as its preceeding rise did not have as long as sustained RSI above 70 as the current correction's run up. The Feb-08 correction was a minor one ith gold falling around 5% (from 942$ to 890$) pretty much in a straight path hitting the bottom after 5 days. It:

• Did cross the 20-day moving average,
• Did not cross or touch the 50 day moving average on its descent to the bottom, 
• Did not touch or cross the lower bollinger band,
• Did just touch the RSI 50.
• Did not cross or touch the RSI 30. 
• There was a period of only one day without a lower-low before the bottom.

The following policy would represent good re-entry points:

• First up day after the RSI touches 50.

The run up to the Dec 09 correction seems the most like the current correction's run up. That is unfortunate as this correction is seemingly the hardest to call a correction for. The Dec-09 correction was a moderate one with gold falling around 12% (from 1222$ to 1078$) jagged path hitting the bottom after 13 days. It:

• Did cross the 20-day moving average,
• Did not cross the 50 day moving average (with a one-day cross back) on its descent to the bottom, 
• Did not touch or cross the lower bollinger band,
• Crossed and stayed below the RSI 50.
• Did not cross or touch the RSI 30. 
• There was a period of three consecutive days without a lower-low before the bottom.

The following policy would represent as good a re-entry point as possible:

• First definite up day after the RSI touches 50.
• Get stopped out on a new low. This happens twice.

Let's take a look at our current condition:

This correction is five days long already and could end tomorrow having crossed the 20 day moving average and with the RSI having touched 50. But only one of five previous cases recovered this quickly. The correction will probably continue at least another couple of days.

After reviewing the five previous cases, my reentry criteria is as follows:

• Reenter at the close of the first real up day after touching or crossing the RSI 50. I'd be more confident if this occurred from below the RSI 50.
• Get stopped out initially at the close of the second day if the closing price is not higher than the previous day. NOTE: Waiting for this confirmation may be desirable (look at the charts yourself) as they eliminate all of the getting stopped out situations except one. I expect I will actually scale in or wait for this second day confirmation.
• Thereafter exit on a lower low for the correction as a whole.
• Repeat as necessary (ouch, this could hurt if we get stopped out repeatedly).
• Where to take profits (after we have them) is for further study.

The Download 20101021gc summarizes the results of this post.

Here's my final observations from the table.

• The May-06 correction was by far the largest and it occurred after the largest runup. That runup was 50% higher than the Oct-10 runup. The difference in runups is evidence that the Oct-10 correction will not be as deep the May-06 correction.
• An approximately 10% correction (which would bring gold down to around $1250) seems pretty likely (see Dec-05, Nov-07 and Dec-09). There's a decent chance a reentry at the end of an up day 10% fall will be stopped out (see May-06 and Dec-09).

The seasonals are another factor that point towards a mild and short correction. October is seasonally bad for gold.

I'm prepared to be stopped out if necessary and I expect I will scale into leveraged long positions at the end of the first decent up-day provided there has been decent price support from India in terms of Indian gold imports in the days preceeding the up day. I'll complete my entering into long positions should the first decent-up day be confirmed the next day. The actual vehicle(s) for going leveraged long are an item for further study.

That's what I'm thinking right now about this gold correction. Nobody knows short-term what is really going to happen. I certainly don't recommend that you act on any of the thoughts in this post. They are just a way of me capturing and sharing my own thinking for the sake of myself and my readers. I do expect I'll be referring to this post frequently as the correction (and recovery thereafter) proceed.

If I have the time I will author a similar study for silver this weekend.

MontyHigh, www.worldofwallstreet.us