Briefly: gold and the rest of the precious metals market are likely to decline in the next several weeks/months and then start another powerful rally. Gold’s strong bullish reversal/rally despite the USD Index’s continuous strength will likely be the signal confirming that the bottom is in.
Welcome to this week's Gold Investment Update.
Predicated on last week’s price moves, our most recently featured medium-term outlook remains the same as the price moves align with our expectations (and we remain on a streak of 11 realized – unleveraged – trades). On that account, there are parts of the previous analysis that didn’t change at all in the earlier days and are written in italics.
The key event of the recent past is still gold’s profound weekly reversal that we saw four weeks ago, which was then supplemented by yet another weekly reversal two weeks ago, and nothing that happened in the meantime invalidated the enormously bearish implications thereof.
In fact, the recent breakdowns in silver (below the head and shoulders pattern), in gold (below its rising support line), and in mining stocks (also below their rising support lines) confirm those bearish indications regarding the short-term.
The USD Index’s breakout provides background, and it confirms that the breakdowns in the precious metals sector were not accidental but rather that they are very likely to be followed by big moves lower.
Today’s run-up is most likely a temporary rebound, and the upside potential seems very limited - I'll explain why in the section dedicated to mining stocks’ short-term performance.
The key fundamental development that happened recently is the Fed’s dovish turn, which is I’ll once again start with a quote from the previous analysis about that topic, and I’ll move to the technical details next.
Debunking Rate-Cut Optimism
The recent narrative uplifting stocks and the PMs is the belief that a pivot is bullish. Conversely, pivots are not bullish, and risk assets often crash when they realize why rate cuts are occurring.
Please see below:
To explain, the blue line above tracks the S&P 500, while the green line tracks the federal funds rate (FFR). If you analyze the horizontal gray lines, you can see that the last three times the Fed cut the FFR, the S&P 500 was already sinking or was approaching a cliff.
Therefore, while it may seem like new highs are inevitable for all assets, the recent optimism is more of a ‘buy the rumor, sell the news’ type trade. In other words, investors will likely bail on the S&P 500 and the GDXJ ETF when the Fed actually cuts rates. To that point, with oil prices resuming their crash, it’s a bad look for global growth when crude oil falls below $70.
Overall, the fundamentals continue to unfold as expected, with higher rates weighing heavily on the U.S. economy. And while the ‘bad news is good news’ trade remains intact, history shows it should end with sharp drawdowns of the S&P 500 and the GDXJ ETF, and a meaningful rise in the USD Index.
Later, I added:
Again, the immediate-term reaction is one thing, as people reacted emotionally to more dovish approach. We also saw an immediate-term rally in 2019, but then a decline materialized, anyway.
So, no, the current move higher doesn’t change the outlook. The big move is still likely to be to the downside. I know that it might be difficult to think so while gold jumped over $30, but this really is the case. Remember how difficult it was to doubt gold’s breakout to new highs? On Dec 4, 2023, I wrote that the breakout is likely to be invalidated and followed by a massive slide. I started that day’s analysis with the following sentence:
“During sharp rallies, it’s nearly impossible to convince investors that this move is about to end. And yet, that’s exactly what is likely.”
That was the top.
And this sentence applies also today. The breakdown below the rising support line in gold was not invalidated and the next big move is likely to be to the downside.
Before I move to gold’s breakdown below its rising support line, let’s consider the key event – the weekly reversal.
I previously wrote about some details coming from the broader perspective and some from the immediate-term analysis, and all that remains true. The recent weekly reversal was so huge that its impact on the next weeks really needs to be kept in mind at all times. It’s not likely that a daily price move in any direction (…) would invalidate the implications of the powerful weekly (!) reversal.
In case that massive reversal seems like a distant memory, here’s a reminder of what happened.
Gold soared above $2,100 only to then plunge to almost $2,000, and in happened on huge volume. This was the most profound weekly reversal in years. It was similar to the reversal that we saw in May 2023 and the one from March 2022, and they both started big declines.
The same is likely this time.
Just like it was the case in early 2023, gold recently corrected to the 61.8% Fibonacci retracement, only to move lower shortly thereafter. This marked the final top before the slide, so the implications here are very bearish as well.
The thing that I would like to emphasize today is that in both past cases, the second week after the final top was the week when gold reversed on an intraweek basis, and this move was not a game-changer.
The above chart shows that this week, gold declined profoundly, but the above chart doesn’t include today’s pre-market upswing. Taking it into account, we would get an intraweek reversal – analogous to what we saw in 2022 and early 2023. Consequently, as history rhymes, today’s move up is likely to be inconsequential. It’s likely just a rebound within a decisive decline that’s going to be followed by more declines.
Zooming in allows us to see that gold broke below two rising support lines and it has now bounced from its 50-day moving average. During yesterday’s intraday rebound, gold once again tested its mid-2023 high (in terms of the daily closing prices), and it then moved back down, closing less than $20 away from the $2,000 mark.
That’s all bearish.
Today’s pre-market performance in gold seems promising, though.
It’s the biggest daily rally in gold that we’ve seen this year, and… That’s as far as its bullishness goes.
Gold is verifying the breakdown below the rising, dashed support line (in terms of the daily closing prices), and while it moved temporarily above it, it’s not likely to close the day and week back above it. And even if the long-term indications are more important, anyway.
What’s really happening, and why it is happening? Yesterday’s session was rich in data – CPI, core CPI, and initial jobless claims statistics were released. This caused some turmoil in the gold market, as it initially gold rose, but then it moved back down, closing the day at new yearly lows. Why would gold be rallying today? Well, gold is a global market, so a large part of the world didn’t have a chance to react to yesterday’s news coming from the U.S. and it’s likely that we’re seeing those reactions today. The very initial reaction in the U.S. was positive, so it’s not that odd that the rest of the world is reacting positively. And, just as the final result was a decline in the U.S. markets, we could see declines globally soon.
Silver moved higher as well, but it didn’t change anything in its particularly bearish technical setup.
As a reminder, silver is after a breakdown below the neck level of its head and shoulders pattern, which has very bearish implications. Today’s pre-market move higher didn’t change the bearish implications at all, as silver topped below the rising neck level of the pattern. Today’s rally, was, therefore another verification of the breakdown.
From the very long-term perspective, we see that silver’s recent top formed at its long-term resistance line. As it held and was not broken, another big move lower is likely to follow.
Breaking below the rising support line (currently at about $22) will serve as a yet another confirmation that MUCH lower silver prices are heading in the following weeks and months.
Given the head-and-shoulders formation as well as the situation in the USD Index and the stock market, the above-mentioned breakdown is highly likely.
Besides, let’s not forget silver’s self-similar pattern.
In early 2021, I wrote that if people wanted to push silver higher, they would need to do it when it was already strong on its own (after a huge decline), and not during a medium-term move lower.
What happened then?
What pretty much always happens after the investment public gets particularly excited about something. We saw a top while silver futures moved slightly above 30 and SLV topped on huge volume – just like it did in 2011.
Looking at historical analogies (note the markings on the above SLV chart) made it obvious that this was a top in the making, and yet, it was nearly impossible to convince people that this huge-volume rally was actually not bullish.
And the historical analogies tell us something very important also this time.
The recent back-and-forth movement in silver is similar to what silver did in late 2012 and early 2013, but that’s just the final part of a long-term analogy.
The final tops (2011, 2021) formed on huge volume, and that was preceded by an initial volume spike. Then the SLV ETF declined in a back-and-forth manner, where we saw 9 bigger highs and lows. Then, after the final (9) top, silver declined in a back-and-forth manner in a smaller range.
We see the same thing right now. The current price movement is what preceded one of the biggest declines in silver’s recent history, so it’s difficult to view the recent performance as something bullish.
The analogy to 2011-2013 had very bearish implications for the following weeks and months, so the implications here are bearish, not bullish.
Yet, just as it was difficult to view silver’s 2021 rally as a top, it’s difficult to view the recent back-and-forth movement in silver as an indication of the upcoming decline – a really big decline.
I wrote about the signals coming from other white precious metals in yesterday’s Gold Trading Alert, but as those implications remain up-to-date, I’ll quote it also in today’s analysis:
Technically speaking, palladium had its run several years ago, and now it’s in a long-term decline, whereas platinum has been weak for many years. In other words, in the recent past, both precious metals were weak.
The above palladium chart shows how important invalidations of major breakouts are. Price tried to move above its 2020 high twice, and once it gave up, it then declined profoundly. It was just only one third of its value at the top.
So, yes, gold’s recent invalidations of breakouts are VERY important and very bearish.
Now, since palladium was so weak, is became a prime candidate for the “weak rally” effect. By that I mean the tendency for the weak markets to soar right before tops. That’s when the investment public is buying, without paying attention to the underlying fundamentals or big trends. People buy just because something looks cheap, without considering that it might be cheap for a good reason.
This causes prices to suddenly spike, often on high volume. We see that effect in silver when it outperforms gold at the tops.
And yes, we also saw in on the palladium market. It jumped up in the previous weeks and it did so on strong volume. It now moved back lower, confirming that the move was not “real”. It was much more like what we saw in early 2022, which was when the precious metals and mining stocks topped.
The implications are – of course – bearish.
What about little silver?
As platinum didn’t enter a long-term bull market in the recent years, it’s not falling back as dramatically as palladium does, but overall, we can still say that this market has been weak in the past decade, especially when compared to gold.
Platinum had a nice short-term run-up recently, and I marked similar run-ups in the recent past. Interestingly, in the vast majority of cases, those rallies preceded important tops not just in platinum, but in the rest of the precious metals market.
The early 2021 is an exception, but it’s perfectly understandable as that was the time right before the Russian invasion and since Russia is the second biggest platinum producer (after South Africa), it’s natural that the markets got concerned with supply issues. In other words, it’s an exception from the rule, not a rule-breaker.
All in all, platinum simply confirms the indications coming from palladium. Namely, that a major top in the precious metals sector has already formed. The same goes for other commodities, as many commodity stocks are pointing to lower prices in the future.
What’s more important than today’s rebound in gold and silver, is what’s going on in the USD Index.
In Wednesday’s Gold Trading Alert, I commented on the USD Index’s short-term performance in the following way:
Ladies and Gentlemen,
The USD Index is breaking higher as I’m writing these words. Precisely, it broke above its declining resistance line, and it’s now verifying the breakout as it moved back to the line that it had broken.
There are two ways in which this is likely to go, and one of them is more likely.
The less likely way, in which this all could develop is that the USD Index fails to break above the resistance line, just like it failed several days ago. In this case, we would likely see another small (to 101.5 or so) move lower, which would then be followed by another consolidation, and then another attempt to break higher. That attempt would be likely to succeed. And if not, then the next one. Either way, if we don’t see a breakout here, what I see as likely is consolidation’s continuation and then a successful breakout. This implies a pause in the precious metals market now, and a slide later (in a week or so).
The more likely way, in which this all could develop is that the USD Index verifies the breakout, and it soars. This is likely not only because we already saw the breakout, but also given the symmetrical nature of the current consolidation relative to the one that we saw in mid-December. The slide leading to the mid-December consolidation was sharp, so the rally that follows the current consolidation, is likely to be sharp as well. This, in turn, means that we won’t have to wait for long to see bigger declines on the precious metals market.
In consequence, the breakdowns that we already saw in some proxies for mining stocks are likely to be followed by bigger slides, and we’re likely to see breakdowns in the remaining proxies (like the HUI Index). Either way, the outlook for the precious metals market (and commodity stocks, like FCX) is very bearish.
The fragment that I put in bold is the scenario that is being realized. The USDX didn’t manage to break higher at its first attempt, but instead of being followed by a decline, the invalidation was followed by another attempt to move higher.
It’s now – after seeing a third attempt to move higher right now – obvious that the USD Index doesn’t want to give up, and that it’s very likely to break out – and hold this breakout – soon. This has profoundly bullish implications for the USDX and profoundly bearish implications for the precious metals sector.
This is in perfect tune with the indications coming from the long-term chart.
The USD Index just invalidated its small breakdown below the rising, long-term support line. This is bullish on its own, and particularly so because the last time it happened, it triggered the biggest rally of the previous year (and last year’s biggest decline in gold and mining stocks).
Also, to make the outlook for the USD Index even more obvious, please keep in mind that the USD Index’s biggest component, the EUR/USD formed a bearish pattern, suggesting a quick reversal – and that’s exactly what we saw.
The European currency moved once again to its early-2023 top, thus likely creating a multi-top pattern. This has been a common way for the EUR/USD to top and I marked the previous cases on the above chart.
In 2008 and in 2011 there were sharp, final comebacks to this level before the decline continued. The one in 2014 was small, but it was still there, and the same goes for the one seen in 2021.
What might see chaotic or a game-changer from the short-term point of view, often becomes normal and orderly when looked at from a bigger perspective.
In fact, the above sentence might be one of the key benefits that professional analyses offer. Thanks to being able to view the bigger picture, and to having a broader perspective (thanks to knowing what to focus on), experienced analysts are able to see the situation as it is, instead of “feeling” what the market is currently “feeling”, which is what most (especially beginning) investors do.
The currency market, gold, and silver are all likely to contribute to lower mining stock prices, and… The miners themselves already indicate that they can’t wait to slide further!
The GDX ETF – proxy for senior mining stocks – moved back above the lower of the rising support lines, but it stopped at the higher of the rising support lines – the dashed one.
The GDXJ ETF – a proxy for junior mining stocks – has done the same thing.
The invalidation of the breakdown below the lower of the lines was likely to be followed a rally, and it happened so quickly – in the first minutes of today’s session – that it wasn’t really tradable.
As the medium-term trend remains down, the really big surprises are likely to be to the downside. For example, if the general stock market declines (and stocks are particularly vulnerable here), junior miners would be likely to plunge.
How high can junior miners rally from here?
Actually, it’s quite possible that their rally has already run its course, or that they will move just a little higher (not higher than $38, and definitely not higher than $40) before sliding again, and then breaking below the rising support line in a decisive manner. With a very limited upside, and significant downside, the outlook remains bearish.
Speaking of stocks, please note that while the S&P 500 and NASDAQ were testing their previous highs, miners are nowhere close to their recent highs. Today’s rally seems to be the “rally of the weak”, meaning that it could be the final day of the strength of the general stock market, and miners – as one of the recently weakest sectors – are showing short-term strength because some people are buying just because the mining stocks are cheap, without considering that they might be cheap for a reason – their medium-term trend, for example. Or the fact that while gold moved above its 2011 highs (I don’t think that it will stay above it for long, before sliding, though), mining stocks are not even close to those highs. In fact, they are well below their 2008 highs.
Also, since the December top, the GDX declined by about 12% (to yesterday’s intraday low), and the GDXJ declined by about 13%.
Junior miners underperform even though the general stock market hasn’t really declined – it’s clear that their downside potential is bigger than the one in senior mining stocks. When the general stock market declines in a more visible manner, junior miners are likely to slide much further.
Now, given today’s upswing in gold, mining stocks are likely to rally as well. However, let’s keep yesterday’s breakdown in mind. Whether this breakdown gets invalidated in daily closing price terms, will determine whether an immediate-term rebound becomes likely.
GDXJ’s medium-term chart shows that it formed a double top, just like it did many times in the past.
The current top is most similar to the 2021 one. Back then, after the price moved below the first op, a sharp slide followed. We already saw the BEGINNING thereof.
This means that my previous comments on the above chart remain up-to-date:
(…) Moreover, please take a look at the areas that I marked with red rectangles. They mark important tops in the GDXJ ETF. In call those cases, junior miners topped by first declining somewhat, then correcting, and then sliding without looking back. In two out of three cases the second (final) top was below the initial one, and in the remaining case (in early 2022), the second (final) top, was slightly higher than the initial one.
So, is seeing the GDXJ close to the previous top (but still below it) a bullish game-changer? Absolutely not.
All in all, the outlook for the precious metals market remains strongly bearish and the potential for our current trading positions in junior mining stocks remains enormous.
What’s next? While the next 1-3 days are a bit unclear, the entire roadmap that I featured for the GDXJ ETF in my previous Gold Trading Alert remains very much up-to-date.
NOTE: Even though I originally wrote the below scenario a few weeks ago, it remains up-to-date.
The markets usually don’t move up or down in a straight line, so some kind of correction is likely to take place in the future, anyway. The question is from what price levels.
My best candidate for the first correction (based on the data that I have available right now) is The $30.5 - $32 range, which is based on the previous lows. I don’t expect a huge rally from those levels, though. Perhaps a move from $30.5 to $32, and then the decline would continue.
The next target is more important. After breaking to new 2023 lows, the move to the 2022 low (close to $26) becomes a good possibility – I marked this area with a green ellipse.
Once this level is reached, I then expect the GDXJ to correct in a more visible manner. After all, at that point, it will be after important breakdowns:
- Below the rising blue support line
- Below the previous 2023 low
- Below the green support line
Consequently, a verification of those breakdowns by a move back to them, would be quite normal. This means a move back to $29 - $30. Then, after a successful verification of those breakdowns, I’d expect the GDXJ to slide lower – to the 2020 low or close to them – at about $20.
There’s also a good possibility of seeing a bottom at about $22, as that’s where we have a downside target based on the head-and-shoulders pattern that is most likely being formed right now. It could also be the case that the GDXJ slides to about $20 on an intraday basis only to recover and close the day at about $22. In a way, both targets would be reached in this case.
There are many IFs around the above-mentioned scenario, and the situation might (and it probably will) change as we go. Remaining open-minded and flexible regarding the new information is key, but having a roadmap is very useful, too, as it shows how things could develop on a more-or-less basis. This can help you prepare for those – or similar – price moves.
Meanwhile, the implications coming from many long-term charts continue to favor lower precious metals and mining stock prices in the following months.
The big news coming from the long-term HUI Index (proxy for gold stocks) chart is the sell signal from the Stochastic indicator. This is not something that we see frequently (we saw it twice last year), but when we do, it’s really time to be paying attention.
This sell signal often follows major tops, and it confirmed the final 2008 and 2012 tops. It doesn’t get much more bearish than this.
In other news, the situation in world stocks hasn’t really improved recently.
The situation continues to be very similar to what we saw in 2008, with the main difference being that now the initial decline was bigger. Because of that it’s no wonder that the corrective upswing is also bigger and it takes longer.
In both cases, world stocks corrected to approximately the 61.8% Fibonacci retracement, and they declined thereafter. This time, there were two corrections to that level, which is not that odd given that the initial decline was bigger this time.
The implications remain very bearish not just for stocks, but also for mining stocks.
As you can see miners (the XAU Index) suffered enormously in 2008 in the aftermath of the situation that’s so similar to the current one. Let’s take a closer look.
This analogy has very bearish implications for the following months.
And while we’re discussing stock market indices and their long-term charts, let’s not forget about the critical situation in the Toronto Stock Exchange Venture Index – one of the proxies for junior mining stocks.
The TSXV is after a massive breakdown below the rising support line, that’s analogous to the one that we saw in early 2013 – and, to a smaller extent, to the one seen in 2008.
Now we see a pause in the decline, but given that the breakdown was already confirmed, the decline is likely to resume after the pause, and it’s likely to put a very bearish pressure on the precious metals and mining stock prices.
Gold stocks to other stocks also clearly show just how important and big the recent “strength” in the mining stocks were.
In short, both are nonexistent. The ratio between the HUI Index (proxy for gold stocks) and the S&P 500 Index (general stock market) is testing its recent lows.
It takes just a little push for the ratio to break below its 2-15, 2018, 2021, and 2022 lows. This is very bearish, because after that breakdown, there’s no significant support all the way down to the 2000 low close to 0.025. This means cutting gold stocks’ prices in half IF the general stock market stays at its current levels, and it means even bigger declines in gold stocks, if other stocks do decline.
Comparing gold stocks to gold reveals that we saw another small correction within a long-term downtrend. Ever since mid-2021, all attempts to break above the 200-day moving average (marked with red) were followed by invalidations and declines in the precious metals sector.
This is bearish, and the same goes for the fact that corrective upswings were smaller in each time when the ratio moved from about 0.1. The 2016 bottom was followed by a sizable correction, the one from 2020 was smaller, then one that we saw in 2022 was even smaller, and this year’s correction is tiny in comparison.
The implications for the following months are very bearish, as when the ratio finally breaks below the rising support line, it’s likely to decline much lower – as low as the 2016 low or even lower.
The stock market recently flashed MAJOR sell signals (I’d view Paul’s decision to take profits from his previous long positions as another bearish confirmation) in the form of invalidations of breakouts to new all-time highs.
The invalidation is clear in the case of the S&P 500 Index futures.
It’s even clearer in the case of the NASDAQ Index, because in this case, this is the first invalidation, not the second one, as we saw in the S&P 500 (in late December).
Both indices moved higher very recently, trying to break to new highs once again, but given how overbought they are and what is the shape of the recent S&P 500 movement, it’s likely that they will fail this attempt and move lower shortly – thus fueling massive declines in junior mining stocks.
The S&P 500 topped in mid-2023 by creating three smaller tops, and we see the same thing right now. Actually, Paul wrote about the likely consolidation here, so what we see is quite normal.
This time, it looks like the S&P 500 (actually, we see the same thing in NASDAQ) is creating a head-and-shoulders top patter and it’s forming the right shoulder thereof. That’s one of the common ways for the markets to top, so the implications are bearish.
As far as tech stocks are concerned, there’s actually more important things going on than the short-term chart shows.
Several decades ago, “tech” stocks were a novelty, and now they are the biggest components of stock market indices. Even the most popular company from the automotive industry – Tesla – is essentially a tech company.
The initial blockchain and AI emotional upswings are over, but it doesn’t make them any less important. Remember the dot-com bubble that followed the all-things-internet bull market? It’s the same thing right now, but with different tech advancements. Even though internet was a game-changer for pretty much all aspects of economy and everyday life, it was also true that the changes did not follow immediately, and that the markets got way ahead of themselves. The same with blockchain and with AI. There were some changes in everyday life, but not tremendous ones – at least not yet.
People’s emotionality remains unchanged despite different geopolitical and technological circumstances, and this also applies to the way in which people react to similar prices moves that were triggered by analogous events. (And markets are created by people, right? Plus algorithms, but those were essentially programmed by humans, anyway.)
Given the above, one would expect that the stock market – at least its tech indices – would behave similarly now to how it behaved around the dot-com peak.
And… That’s exactly what’s happening.
The above chart features the NASDAQ (main line) as well as gold and silver stocks (the XAU Index serves as a proxy; orange line).
There is only one period comparable to the recent over-decade-long rally and that’s the huge rally that preceded the 2000 top. It’s not like we have to look for the analogy – clear crystal-clear when looking at the chart for just a second.
I already described the underlying technical changes that triggered both upswings, so I want to focus on something else now. Namely, on the fact that since the long-term momentum was already broken, it’s obvious that the moment that is analogous to what we see right now is what we saw in 2000 after the top.
That was when the momentum in the previous long-term upswing was broken, and the NASDAQ moved below its 50-week moving average (marked with blue). It then moved back up again, but that was not the continuation of the uptrend, even though it looked like it. In reality, it was the first correction after the start of a massive decline.
This is most likely where we see tech stocks right now as well. We saw the distinct slide below the 50-week moving average, and now we experienced a short-term rally. Some will say that this is just the continuation of the previous long-term uptrend, but the historical analogy suggests otherwise. It seems much more like the first correction after the start of a massive decline.
Now, if tech stocks are about to turn south and then decline in really substantial manner, what does it imply for the precious metals market?
There are times when tech stocks and the precious metals stocks move in the same direction, and there are times when they move in the opposite directions. For example, they plunged together in 2008 and in 2020. This is one of those times, and you can see details in the bottom part of the above chart.
The correlation coefficient that’s visible there takes values from -1 to 1, and the closer it is to -1, the more markets tend to move in the opposite directions, the closer to 1 it is, the more markets move in the same direction, and the closer to 0 it is, the more markets move regardless of each other.
At 0.69, the coefficient indicates a situation, in which both markets move in the same direction.
This means that a big move lower in the NASDAQ is likely to translate into a big move lower in other stocks and also in mining stocks. Quite likely also in copper.
And by bigger, I don’t mean “just big”. Based on the past patterns in the tech stocks, it seems that a 2008-style decline – or worse – is to be expected. This means a huge decline in mining stocks is likely around the corner.
Truth be told, such a bearish storm for mining stocks has been brewing for some time. It’s clear once you look at the correlation coefficient itself and focus on the times when it’s been rising for a long time, and then it started to decline.
I marked the rallies (usually from the 0 level) in the correlation with orange rectangles. In each of the previous cases, when the correlation peaked, it meant that the XAU Index was about to move significantly lower. As you can see on the above chart, the correlation peaked in 2021 and while the XAU declined since that time, the size of the decline is not yet comparable to what used to happen after similar signals.
This means that gold and silver mining stocks have likely not declined enough just yet. Therefore, they are likely to slide much more in the following months. This move lower started in 2022, and given the recent short-term events, it seems that it’s about to continue. There will be short-term trading opportunities in both directions, but the underlying big trend in the mining stocks is currently down. And if that’s not clear, please consider the fact that while gold is well above its 2008 highs, gold and silver stocks, are well below that high. In fact, they are even below their 2006 high.
Speaking of copper, the situation in it also confirms that a turnaround is most likely upon us.
The recent rally was quite sharp, but it took copper back to its rising resistance line, which is also the neck level of the head-and-shoulders formation. A decline from here would be a natural consequence of the verification of the breakdown. The target based on the head-and-shoulders formation is below 3.0, and that’s where copper is likely heading, at least initially.
The short-term picture for copper is also bearish (although rather unclear), and you’ll find details below in the quote from yesterday’s Oil Trading Alert (while there’s just “Oil” in the name of the service, it covers many commodities like natural gas and copper as well as individual oil stocks; on a side note, Anna’s current trade in crude oil is profitable, with a good short-term potential):
===
Copper - The Current Overview
In my Oil Trading Alert published a week ago, you could read the following:
(…) the green dashed rising support line didn’t manage to stop the sellers, and Tuesday’s daily closure took place under this support. This show of the bulls’ weakness encouraged their opponents to act and resulted in a red gap (3.872-3.881) at the beginning of the next session.
Although the buyers tried to close it, they failed quickly, which translated into another downswing that pushed the price to our next downside target – the Dec. 14 low and the green gap (3.788-3.835) formed that day.
The volume, which was accompanied by recent declines was increasing from session to session, confirming the sellers' involvement in shaping the next red bearish candles. Additionally, the CCI and the Stochastic Oscillator generated sell signal, giving the bears even more reasons to act in the following days.
If this is the case and the commodity extends losses from here, the first downside target for the sellers would be around 3,80, where the 38.2% Fibonacci retracement (based on the entire October-December upward move) and the 200-day moving average currently are.
Slightly below these supports is also the lower border of the green gap formed on Dec.14 (3.788-3.835) and the gap (3.789-3.829) formed at the end of November (marked on the weekly chart), which together serve as the major short-term support, which was strong enough to stop declines in December.
(…) What could happen if the bears manage to close both green price gaps?
It would be a strong bearish signal, which will likely open the way to the 50% Fibonacci retracement (around 3.75) and the early December lows (at around 3.731).
Looking at the daily chart from today’s point of view, we see that the situation developed according to the mentioned scenario, and copper declined to the next downside target.
Before we focus on the meaning of recent events and their significance for the future fate of the price, let's analyze together what happened on the way to current levels.
On the day of publication of the last comment, the bulls attempted to close the red gap created the day before. Their actions, however, had no effect and after a slight increase after the opening of the session, the price dropped below the opening, which opened the way for the bears to the south.
In the following days, the declines deepened, and the copper tested the lower border of the green gap (3.788-3.835), which served as the main support in the short term. Although the buyers pushed the price higher, this improvement was only temporary and the next session brought sharp decline, which resulted not only in a pro-declining formation (the bearish engulfing pattern – marked with the red ellipse), but also in a closure of the gap and a decline below the 50- and 200-day moving averages.
This combination of negative developments resulted in a drop to the next green area created by the Dec. lows and the 50% Fibonacci retracement, which serves as the nearest support.
What’s next?
All the above-mentioned bearish developments remain in the cars, suggesting that recovering lost higher levels won't be easy – especially when we factor in the recent volume, which was visibly lower during recent upswings (both white candles were created on a lower volume than the red ones, which emphasizes the involvement of bears in price formation).
Nevertheless, the proximity to the green support zone and the current position of the daily indicators (they slipped to their oversold areas, which suggests that buy signals may be just around the corner) suggests that the space for declines may be limited and reversal in the coming days can’t be ruled out.
Finishing today’s alert, please keep in mind that a positive scenario will be even more likely if the bulls manage to push the price higher before the end of the week and close it above the green gap marked on the weekly chart.
What could happen if they fail?
The next stop could be the previously broken upper border of the red declining trend channel (marked with red dashed line on the daily chart around 3.72) or the 61.8% Fibonacci retracement (around 3.70) based on the entire Oct-Dec. upward move.
Summing up, the bulls failed to stop their opponents in the previous week, which resulted in a loss of valuable allies (the green gap and two moving averages) and a drop to the next support area. Its proximity with the combination with the current position of the daily indicators suggests that reversal may be just around the corner.
===
Naturally, the above would be likely to have a very negative impact on the prices of copper stocks, like FCX.
The FCX just invalidated its move above the declining dashed resistance line, and when that happened in August 2023, it meant that the top was in. The implications here are – of course – bearish.
Since the price target for copper is well below its current price, FCX is likely to fall far from the current levels as well.
The declines from the current levels after failed breakouts above the 2007 high were big, and in some cases (in 2008) they were enormous. The downside potential is waaaay bigger than what you can see on FCX’s short-term chart.
Finally, the analysis of sentiment continues to support lower gold prices in the following weeks and months. My previous sentiment analysis remains up-to-date:
Have you considered buying gold recently? Like to the point of searching for it online?
Because many people have.
As you can see on the above Google Trends screenshot, the searches for “how to buy gold” just soared, and it’s not the first time that it happened.
Makes one wonder… What happened to gold price in those other cases?
After all, whatever circumstances triggered this jump in the interest in the topic, they are taking place all over again. I don’t mean the state the world is in – I mean the sentiment among gold investors. By estimating the latter, we can also estimate what’s likely to happen to the price, because… The history tends to rhyme, and people’s emotional reactions to what the market is doing remain more or less the same, regardless of the details of the fundamental situation.
I marked one of the moments on the above chart and here are the other notable peaks:
So, what happened to gold price on those occasions?
I marked all-above-mentioned cases with blue, dashed lines and in three out of four cases those were the MAJOR tops. Ones that were followed by hundreds-of-dollar declines in the price of gold.
The only remaining case was when it was still the end of a short-term rally and the start of a pause (that took gold about $100 lower, anyway). This time was truly exceptional, though, because it was right after the covid-scare bottom – it was not a regular course of action.
So, I’d say that in all “regular” cases, the huge increase in interest in buying gold translated into huge declines in the following months. After all, people tend to buy at the tops – that’s exactly what this sentiment analysis proves.
We are at this stage one more time (in many other stocks, too, including some oil stocks). Once again, it’s difficult NOT to buy into the euphoria, even though looking at the situation from a broad perspective practically “screams” WATCH OUT.
Now, you are informed, you are prepared.
And I will continue to keep you – my subscribers – up-to-date, so that what surprises most investors, will not surprise you, but that it will benefit you. We’re on a streak of 11 profitable (unleveraged) trades, after all, and it’s VERY likely that the current trades will increase this streak soon.
Again, as always, I’ll keep you – my subscribers – updated.
= = =
If you’d like to become a partner/investor in Golden Meadow, you’ll find more details in the above link.
Overview of the Upcoming Part of the Decline
- It seems that the recent – and probably final – corrective upswing in the precious metals sector is over.
- If we see a situation where miners slide in a meaningful and volatile way while silver doesn’t (it just declines moderately), I plan to – once again – switch from short positions in miners to short positions in silver. At this time, it’s too early to say at what price levels this could take place and if we get this kind of opportunity at all.
- I plan to switch from the short positions in junior mining stocks or silver (whichever I’ll have at that moment) to long positions in junior mining stocks when gold / mining stocks move to their 2020 lows (approximately). While I’m probably not going to write about it at this stage yet, this is when some investors might consider getting back in with their long-term investing capital (or perhaps 1/3 or 1/2 thereof).
- I plan to return to short positions in junior mining stocks after a rebound – and the rebound could take gold from about $1,450 to about $1,550, and it could take the GDXJ from about $20 to about $24. In other words, I’m currently planning to go long when GDXJ is close to $20 (which might take place when gold is close to $1,450), and I’m planning to exit this long position and re-enter the short position once we see a corrective rally to $24 in the GDXJ (which might take place when gold is close to $1,550).
- I plan to exit all remaining short positions once gold shows substantial strength relative to the USD Index while the latter is still rallying. This may be the case with gold prices close to $1,400 and GDXJ close to $15 . This moment (when gold performs very strongly against the rallying USD and miners are strong relative to gold after its substantial decline) is likely to be the best entry point for long-term investments, in my view. This can also happen with gold close to $1,400, but at the moment it’s too early to say with certainty.
- The above is based on the information available today, and it might change in the following days/weeks.
You will find my general overview of the outlook for gold on the chart below:
Please note that the above timing details are relatively broad and “for general overview only” – so that you know more or less what I think and how volatile I think the moves are likely to be – on an approximate basis. These time targets are not binding nor clear enough for me to think that they should be used for purchasing options, warrants, or similar instruments.
Letters to the Editor
Depending on the nature and target group of your question, feedback or comment, please use the following means of communication:
- For questions or comments that you’d like to get the Community’s response, please use the Ask the Community space so others can contribute to the reply and also enjoy the answers.
- For questions, comments or feedback that you’d like me to comment on / reply to, please send them to Golden Meadow’s support – some clarifications can be provided directly by our experienced support team, and those that are strictly for me, will be forwarded to me and I’ll then provide replies either individually, or in the “Letters to the Editor” section in the Alerts, depending on the nature of the question/comment.
Administrative Announcement
The price of the Gold Investment Updates subscription is going to be increased on January 18, 2024, OR on February 18, 2024.
The prices of pretty much everything have gone up in the recent months and years, but the price of Gold Investment Updates service hasn’t been updated a single time, while at the same time the scope of the analysis continued to increase (at its inception, it didn’t include copper, for example) and every now and then we’re adding extra value by providing intraday messages and sometimes we’re even providing access to Gold Trading Alerts for several days (that are currently valued at $299 per month with discounts available for yearly subscriptions). As or our knowledge Gold Investment Updates also the most comprehensive weekly publication covering gold, silver, mining stocks AND related markets that’s available.
The new monthly price is $99 per month (with discounts available for yearly subscriptions).
The time when the price comes into effect depends on when one’s subscription will be moved to Golden Meadow. As we are retiring our old website (SunshineProfits.com), it’s best if everyone moves to the new platform (GoldenMeadow.eu) soon. To make it more convenient for everyone, we are delaying the price increase by one month to everyone that moves (or has already moved) their subscription to Golden Meadow.
Our support team will be reaching out to every Gold Investment Updates subscriber (a few people already moved to Golden Meadow – in this case, no further action is necessary, thank you for being so quick!) with a special link that properly adds the already paid-for days to your Golden Meadow subscription and ensures that the price increase will affect with you a monthly delay. In other words, you don’t need to do anything before receiving a link from our support team.
If you’d like to use this opportunity to upgrade your subscription to Gold Trading Alerts, or maybe even to the prestigious Diamon Package, please let our support team know – they will make sure that your upgrade is successful.
It looks like this year is going to be full of very important developments not just on the precious metals market, but on all markets, including stocks and currencies. Thank you for staying updated with our premium services in those times.
Thank you.
Summary
To summarize, today’s quick run-up in gold, silver, and miners is likely just a bounce within a bigger downtrend. Gold is after two major weekly reversals, which point to MUCH lower prices in the following weeks. The same goes for silver’s confirmed breakdown below its head and shoulder’s pattern, and USD Index’s breakout above its medium-term declining resistance line, not to mention the peak in interest in “how to buy gold” searches.
The enormous potential of our trading position in the junior miners remains intact.
Also, given that it’s highly unlikely that we’ll see any major rally here, I’m adding stop-loss levels to all our positions, and I’m adjusting the target levels for the leveraged instruments as well.
Also, given where the RSI is right now for the GDXJ, and considering today’s rebound (it’s the first notable correction within this decline), it became more likely that we’ll get a rebound from the 2023 lows, before the GDXJ moves lower.
Consequently, I’m moving the target prices closer to the current levels, to focus on a more short-term trade.
This does NOT imply a smaller overall profitability, nor does it imply that I’m expecting to see a smaller decline in general. I’m simply seeing a good possibility of seeing a bigger rebound earlier, and then a bigger decline as a follow-up action. This is likely to increase our overall profitability from the medium-term decline, as we’re likely to have more profitable trades this way.
===
As we’re on a streak of 11 profitable (closed, unleveraged) trades, and – just like I wrote today and in the previous days – it looks like we’re going to see much more of them in the near future, I want to provide you with even more great news!
There are even more savings connected with our Diamond Package that includes Gold Trading Alerts, as well as Oil Trading Alerts and Stock Trading Alerts.
There are many reasons due to which the above are interesting, for example, the recent huge profits that Paul took off the table in Stock Trading Alerts, and Anna’s already-profitable (with great short-term potential) trade in crude oil in Oil Trading Alerts.
The prestigious Diamond Package includes a bundle-offer discount, there’s a 10% discount available for purchasing it for more than one year, and on top of that when using the “DIAMOND10” code when going Diamond there’s an additional 10% discount on top of it all.
Additionally, if you’re interested in trying out Oil Trading Alerts and/or Stock Trading Alerts, please note that there’s a 7-day free trial available for both services. You can sign up for the free trials (or go Diamond) using this page.
Please contact our support to upgrade your subscription to ensure that your paid-for days will be properly transferred over to your new subscription.
To summarize:
Short-term outlook for the precious metals sector (our opinion on the next 1-6 weeks): Bearish
Medium-term outlook for the precious metals sector (our opinion for the period between 1.5 and 6 months): Bearish initially, then possibly Bullish
Long-term outlook for the precious metals sector (our opinion for the period between 6 and 24 months from now): Bullish
Very long-term outlook for the precious metals sector (our opinion for the period starting 2 years from now): Bullish
As a reminder, Gold Investment Updates are posted approximately once per week. We are usually posting them on Monday, but we can’t promise that it will be the case each week.
Our preferred ways to invest in and to trade gold along with the reasoning can be found in the how to buy gold section. Additionally, our preferred ETFs and ETNs can be found in our Gold & Silver ETF Ranking.
Moreover, Gold & Silver Trading Alerts are posted before or on each trading day (we usually post them before the opening bell, but we don’t promise doing that each day). If there’s anything urgent, we will send you an additional small alert before posting the main one.
Thank you.
Przemyslaw K. Radomski, CFA
Founder, Editor-in-chief