Please read the Disclaimer before reading this article.
Short summary
We saw very constructive price action these last two weeks in the way where gold bounced and in the way the RSI behaved.
In my opinion, there is a 50% chance we have seen the low, 40% chance we go lower again (below 1760) but should see a strong close of the month (this is vital), and a 10% chance that something more bad happens (since I don't have a clear view on the short-term fundamentals, this option needs to remain open)
Introduction
In my last article, I was wondering if we were going to see that final move down in gold to the 1800 level that I have been envisaging since early August.
This move started to happen the day after I wrote that article.
This gave me the opportunity to add the second half of my capital into the mining shares, as shown below:
(Please note that I use GDX as a reference because I use it as my timing device for GDXJ and other mining shares.)
After I bought, gold (on an intraday basis) declined another 40 dollars but mining shares showed relative strength and didn’t continue to decline.
In this article, I want to explain why I bought that second position and went all-in.
I’ll do this by:
-showing the historical price action of gold in the US Dollar and other currencies when they reach new all-time highs.
-showing the clear formation that was visible on the daily RSI for gold.
Historical analysis of all-time highs
As I’ve explained before in several articles, I was expecting the price action in gold in USD to follow the same path as gold did in other currencies like the Euro and GBP.
The following chart shows that this is exactly what happened. Gold reached this week the support level and we saw a strong bounce, indicating the importance of this level.
Several analysts are seeing this recent rally as a short-term bounce and are indicating lower price targets (in the low 1700’s and high 1600’s) are likely ahead. For me, these lower price levels are difficult to envisage because as I’ll show, historically, gold has always shown the same pattern when it broke to new all-time highs.
(As I'll explain at the end of the article, lower prices are perfectly possible but because of my trade management, it doesn't matter for me.)
The pattern is as follows:
Breakout above all-time highs -> sharp rally -> decline to retest the breakout -> new sharp rally.
We saw this pattern playing out for gold in both the Euro and the GBP.
If we look at the historical price action of gold in the USD, we see the exact same thing.
When Nixon ended the link between gold and the US Dollar in 1971, gold prices rose sharply and peaked (temporarily) in early 1975.
We saw a correction that lasted a couple of years and in 1978, gold was able to break above those former all-time highs that were set in 1975.
After the breakout followed a sharp rally, followed by a retest from the breakout, followed again by a sharp rally.
The exact same pattern as we have seen the last few years for gold in the Euro and GBP.
The gold market peaked in 1980 and so this was the new all-time high that needed to be taken out. This happened in September 2007.
You can clearly see the consolidation in the gold price that happened in the months before the breakout (using monthly closing prices), indicating the significance of this level.
Once the breakout happened, the exact same pattern played out again: strong rally -> sharp decline to retest the breakout -> sharp rally.
The peak that was established in March 2008 was the next all-time high that needed to be taken out, once prices had retested the breakout level.
As the following chart shows, once again, the pattern was exactly the same…
So, while some analysts would not be bothered by gold prices going to the low 1700 level, my historical analysis shows me that for the investment framework that I follow, I wouldn’t like it at all.
This historical insight was the first reason why I bought the second half of my position. If gold is still in a bull market, the low should have been very close when I bought my second position.
Daily RSI analysis
I bought gold mining shares in March 2020 and sold them at the end of July/beginning of August. I stayed bearish during the whole correction and initiated my first buy 3 weeks ago and my second buy 2 weeks ago, as indicated at the beginning of this article.
One of the primary reasons I stayed bearish, was the lack of a clear bottoming formation, which was clearly visible at the end of the 2 previous corrections in this gold bull market (see chart below).
One of my followers asked me a few weeks ago the following: if gold reaches the 1800 level, would we need to wait for a clear bottoming formation to show itself before we get in again?
I answered that it depends on the type of decline. If it is a normal decline, I would like to see a bottoming formation around the red horizontal line on the RSI chart (see below).
However, if gold declined sharply, what I would absolutely not want to see is an oversold RSI.
Even during the liquidity event in March, where gold got completely crushed, the RSI did not get oversold.
And now, once again, the RSI did not get oversold, this is a very bullish signal (as long as it stays this way of course).
Why is this bullish? Let me explain below.
In an article I wrote on October 11, I said the following: “We all have access to the same data. The trick is to interpret this data correctly. I think this is one of the biggest mistake investors make. They just think an overbought RSI is bearish and an oversold RSI is bullish.
In my previous article I used the extreme RSI indeed as a sell signal. However, most of the time I use an overbought RSI as a sign we haven’t seen the top yet.
The trick is to understand that your interpretation of a certain indicator is dependent on the regime the market is in. In future articles this will become clearer.”
A lot of people see an oversold RSI as a sign that prices will bounce sharply. However, as explained in my quote above, it all depends on the market regime.
An oversold RSI in a bull market is a bearish sign, not a bullish sign.
On the following chart, I indicated all the RSI oversold signals with a red vertical line.
As you can see, all these red lines took place during the multi-year basing pattern in gold. Once gold bottomed in September 2018 and broke out in June 2019, we haven’t seen any oversold signal…
As the chart indicates, an oversold signal can mean one of two things:
-we either get a sharp rally but we are probably not in a bull market and prices will decline again after this (see the instances in 2013 and 2014)
-or prices continue to decline. The oversold RSI is indicating the weakness of the market; it is not indicating a buying opportunity. (see the instances in the second half of 2016 and 2018)
If we look at the bull market in gold after the Great Financial Crisis, we see exactly the same pattern: prices never got oversold during the bull market, and once they got oversold, we had already seen the peak in gold and it correctly indicated that the nature of the market had changed. The bull market was over and a long term topping pattern was starting to form.
So, to put it together, why did I buy my second position in the mining shares?
I knew from a historical perspective that prices should not drop below the previous all-time highs and I knew that if we were in a bull market, the RSI should not get oversold.
When I bought, the RSI was just a tiny bit above the 30 level. On Monday, when gold declined intraday below 1770, the RSI bottomed exactly at the 30 level…
So, gold bottomed exactly at the level it needed to, with an RSI pattern that supports the continuation of the bull market.
It seems that the dream scenario I envisaged on November 8th is taking place: “In my previous article, I explained the dream scenario I was envisaging. The dollar would rise and gold would decline to the 1800 level. At the same time, the mining shares would show relative strength and a clear bottoming formation would establish itself.”
We saw the move to the 1800 level, we saw relative strength in the mining shares, and we saw a clear bottoming formation (an RSI that bottomed at the 30 level instead of a positive divergence). The only thing that we didn’t see was a strong dollar. It is interesting to note that the decline in gold from 1950 to 1770 dollars, happened when both the dollar and real interest rates were declining…
Conclusion
I use a combination of sentiment, technical, and fundamental analysis. Both the sentiment and technical picture are now perfect for gold.
While I’m convinced about the long-term bullish case for gold, the short term fundamental picture is still not clear to me. How will I manage my position?
If this was indeed the low, I’ll be happy and make big gains the coming months, nothing to do.
If we go lower again, this doesn’t really bother me. Because I bought my full position at two different prices, my average price is below the current price, meaning I currently have a profit. If I use a strict stop order, it is impossible for me to lose money on this trade.
In practice, I’ll analyze the price action in gold the following days to see what I’ll do. On Wednesday, I posted on my twitter account the short term levels I’m watching. If we decline below 1819 again, I’ll probably sell my whole position. Does this mean that I’ll then become bearish? No, it is just my trade management. My loss will be minimal and I’ll be able to analyze what the market will do. I’ll have my full capital ready to buy again. Or, in the worst case scenario, when gold indeed goes to much lower levels and the RSI gets oversold, I will lose almost nothing.
You see, if your trade execution is good, meaning that you didn’t get sucked in by all those people who have been calling a bottom and breakout when gold was still above 1860 and even 1920, you shouldn’t worry too much.
My analysis can now be completely wrong, gold can go to 1000 dollars, and I’ll lose almost nothing…
One last important side note. If you analyze the historical charts above that show the price action in gold once new all-time highs were broken, you’ll see that I sometimes use daily, sometimes weekly, and sometimes monthly closing prices. This indicates that gold can decline to the low 1700 level, as long as it closes the month above the support level. In this instance, I will probably have sold my full position and be in cash. If I see the gold price in the low 1700’s and the RSI is not oversold, I’ll probably buy again since probabilities are that the month will close above 1780. If gold declines to the low 1700's but the mining shares show relative strength and don't decline, I'll probably stay with them.
(If you look at the other times when gold bottomed with an RSI of 30, in some instances that was the bottom, in other instances, we saw a rally and thereafter a slightly lower low on a positive divergence. We'll have to wait and see. In both cases, I'll know what to do. And that is the important lesson. It's not about timing the bottom perfectly to the day. It's about timing the bottom better than most other people so that you can have a trading plan that can minimize losses and maximize gains.)
I’ll update my short term views more frequently on my twitter account. So, please follow me there.
Starting at the end of next week, I’ll also have more time free to write more frequently. This will allow me to touch on other topics like the S&P500, as some readers have asked for.
Thank you very much for reading my article!
Comments