jueves, 13 de julio de 2017

jueves, julio 13, 2017

Gold: Have We Hit Bottom Yet?

Taylor Dart


 
- Bullish sentiment on gold is trading at the lowest levels since December.
       
- The metal has plummeted $90/oz over the past 20 trading sessions.
       
- COT Data has finally begun to swing positive for gold, at is at levels consistent with where bottoms begin to form.

       
The price of gold (GLD) has been pummeled the past 6 weeks with seemingly no bottom in sight. Bullish sentiment has plunged to the lowest levels in 7 months, and miners are now trading at more attractive valuations than the Christmas lows. The million dollar question is whether it's time to step in now and take a stab, or if there's still more downside left to be seen. Bullish sentiment is at depressed levels and COT data has swung favorable, but price itself (the most important indicator) has not reached support just yet. Add this to the fact that 5-week losing streaks have always seen further draw-downs, and the best play seems to be patience at this juncture.
 
(Source: TC2000.com)

A couple weeks ago I put out a note on gold and stated that the metal had a setup in place from which most of the ingredients for a bounce were there, but the key was defending $1,240/oz. Failure to hold the $1,240/oz level would not be a good omen for the bulls, as there was a pocket of air below down to $1,200/oz. This provided a low-risk long setup to get long vs. $1,245/oz with a stop below $1,240/oz. While the metal did bounce shortly after, it ran right into resistance at $1,260/oz. Shortly after, the $1,240/oz level was violated, and the rest is history.
 
It's very rare that I say that bad news is good news, but in terms of the gold market here, this may be one of those times. I was really hoping that we would see an ugly week for gold last week, to allow us to register a fairly rare signal, and fortunately, we've got it. Gold ended last week in the red and in doing so recorded its fifth straight weekly decline, something we've only seen 11 times in the past 18 years. The results after these signals are quite interesting and tend to be quite positive for gold looking one month ahead. The caveat is that every single time we've seen a 5-week losing streak, the close of the 5th week has never marked the bottom, and there's always been some more pain ahead. Let's dig a little deeper into the study....
 
Five-Week Losing Streaks
 
(Source: Microsoft Excel, Author's Table)


Since 1999, we've only had eleven 5-week losing streaks for gold, and we have been higher a month later 73% of the time by an average of 2.1%. This means that if we were to perform with the average of this signal, gold would end the first week of August at the $1,235/oz area. The smallest decline over the next month we've seen after the fifth straight losing week was 0.73%, with the worst being 6.31% - the average decline: 2.74%. These numbers are intra-day numbers as they are draw-downs, so this is not returns, but just where gold has traded down to after recording 5 straight losing weeks in the past.

 
(Source: TC2000.com)
Given that we closed the 5th straight losing week at $1,210/oz, if we were to perform in line with the average decline, we'd see ourselves touching $1,176/oz between now and August 7th.

It's worth noting that even if we do go that low, the average signal manages to finish the month with a 2.1% return. What this means is that there's a setup here where the average decline is 2.74% over the next month, with the month finishing with a 2.1% return on average. The strategy is pretty simple, if we pullback into the $1,180 - $1,192/oz area, it's likely a nice risk/reward spot to buy the dip.
 
As for draw-ups, we have always traded higher than the 5th straight losing week's close at some point, and the average move higher (draw-up) over the next month has been 4.80%. This means that if we were to trade with the average, at some time over the next month we would touch $1,268/oz. Certainly the opposite of what most are expecting right now.
 
Forward returns on a 1-week basis tend to be poor, with 1-week returns averaging 0.17%, but only seeing a positive return 36% of the time. Having said that forward 1-month returns are much better as explained above. So if we were to trade in line with past occurrences, we've likely got another week or so of pain ahead, but can expect to begin the month of August higher than where we are now. The ideal setup would be for us to be red this week, as past occurrences when we get a sixth straight losing week are higher 1-month later 86% of the time. The worst thing possible according to this signal is a strong week this week and a large green bar, the only 2 times we saw this, we saw the worst 1-month returns going forward.
 
Past instances and returns do not guarantee future returns and performance, but buying the dip if we can see a pullback into $1,180/oz - $1,192/oz seems like a decent setup here, and the bulls likely have a good shot at turning this around very soon. The key is there doesn't seem to be any rush to jump in just yet to buy either miners or gold, a better opportunity should lie ahead during the next two weeks.
(Source: TC2000.com)

 
To avoid any confusion, forward returns in the table are based on if gold was bought at the close of the fifth straight losing week, the numbers are not weekly/monthly performance numbers, they are where the return would stand if we had bought the close of the 5th straight down week.
 
COT Data
 
Moving onto COT data, for the first time since the December lows the commercials are at a very modest net short position, and this was as of mid last week while we still sat near the $1,220/oz level. When looking at COT data, the most important indicator to me is what the commercials are doing as they have been the most reliable indicator over the past couple decades when it comes to tops and bottoms in commodities. Having said that, I do pay attention to what small speculators are doing as they've been known also to be good contarian signals when used in conjunction with commercials positioning.
 
(Source: GoldSeek.com)
(Source: TC2000.com)


As we can from the above charts, commercials are now less net short than they were at the December lows post-Fed, while small speculators are significantly less net long than they were at the December lows for gold. This means that commercials who I want to see paring back their net short position have done so dramatically, and small speculators want absolutely nothing to do with gold right now as they are nearly flat on their positioning - something we very rarely see. For those that have difficulty making sense of the chart, the commercials were net short 150,000 contracts in December, and are currently only net short 107,000 contracts. Meanwhile, the small speculators were net long 20,500 contracts at the December lows, and are currently net long significantly less with 13,000~ contacts.

Does this mean we have to bottom? Absolutely not. But we are at levels in the COT data that are consistent with previous bottoms, so the box can be checked for this indicator being a "go".
 
Technicals
 
Moving onto the technicals, the key here is if we still have support below and are still in a bull market. Depressed sentiment can be a good indicator to time bottoms, but it is much more reliable to see depressed sentiment in a bull market vs. a bear market. Indicators used to time the long side become much less reliable in bear markets as absolute panic takes over and things can go much further than anticipated. The same is true of using indicators to time the short side in euphoric bull markets, just ask those that tried to short in the Dot Com era. While a few made out well trying to bet against a market that had asinine valuations, the majority failed miserably.
 
(Source: TC2000.com)
 
 
Taking a look at the monthly chart of gold, we can see that we are currently violating the 20-month moving average which is not a particularly positive development for the bulls. Having said that, I only care about this indicator when it comes to monthly closes, so this current violation is meaningless unless we close the month below here. The 20-month moving average currently sits at roughly $1,240/oz, so the bulls have some serious work to do to avoid this. As we can see gold also broke below the 20-month moving average near the December lows and this was not a deal-breaker for the metal. The 20-month moving average is simply a gauge for me to assess the health of a market, and the more months spent the 20-month moving average, the more likely a bear is to rear its ugly head. The more important takeaway from this chart is the uptrend line off of the 2015 bottom that I would not like to see violated. This uptrend line currently comes in around the $1,180/oz level, and also coincides with horizontal support.
(Source: TC2000.com)
 
 
Moving to a weekly chart we can get a better look at this horizontal support level and the weekly higher swing lows we had seen up until recently. While we did violate this pattern and make a lower swing low vs. the May one recently, the key for me is defending the January and March weekly swing lows at $1,179/oz and $1,194/oz. It is possible that we violate these levels to the downside intra-day, but what I'll be watching for is where we finish the day and week. A daily and or weekly close below $1,180/oz would be a massive dent in my bullish thesis and would force me to exit my half position in gold that I put on at $1,176/oz last year.
 
So how am I positioned?
 
Once $1,240/oz broke I began to get defensive in my miner allocation and pared back from a 35% weighting in miners to 22% as of most recently. This is because one of my risk management rules triggered that only allows me a certain amount of exposure to an industry that is ranked in the bottom 20% of all industries. The early month move in gold and gold stocks triggered this rule, and I was forced to liquidate quite a few positions. Having said that I have a few names, and have a shopping list of undervalued miners I'm waiting to deploy in my premium newsletter.
 
My top 5 weighted miner positions in no particular order are:
Alio Gold remains the most attractive from a valuation standpoint of the current miners I hold along with Semafo Gold (OTCPK:SEMFF), NewCastle Gold and Marathon Gold are the most attractive from a take-over standpoint among the juniors with undervalued premium ounces in favorable jurisdictions, and Osisko is my the most de-risked junior I own and one of the only juniors I see with the potential to prove up 10 million plus ounces across their properties company-wide.

In summary, I see no reason to jump on the buy button yet for those that already have exposure, as we have always seen weakness after 5-week losing streaks. If I did not have any exposure whatsoever here, I would begin nibbling and starting positions in a couple names with room to add when I'm confident a bottom is in.
 
Gold trades on sentiment as much as it does on technicals and the metal is always pushed to extremes in both directions. Based on the average draw-down from 5-week losing streak signals, at the minimum, we should see gold trade down to $1,201/oz before it finds its footing.
 
It is entirely possible that gold does not pullback with the average decline of 2.74% after these signals and instead makes a higher low. I am open to seeing a higher low for gold, but am cautious of this as the majority of the time we do see more pain and to the tune of at least 2%. If we do make a higher low, this would be a positive for the bulls and I'll be looking for certain indications on my swing charts to tell me that there's a high-probability the bottom is in.
I'm likely going to sound like a complete broken record by now, but what does this all mean? If we do see a 1-2% pullback over the next couple weeks, we'll be pulling back into my support zone between $1,180/oz and $1,200/oz, we'll be doing so with sentiment in the dumps at an area where it usually bottoms out (at least for a short term bounce), at an area where Markos has turned bearish and finally thrown in the towel on the metal, and we have forward returns strongly in our favor after this setup has occurred in the past. There are no guarantees in life and certainly not in the market, but this would likely be as close to a perfect storm as we can get for a low-risk long entry.
 
As stated above, due to the fact that I have a decent amount of long exposure I'm sitting on my hands for the time being. If I had no long exposure or minimal long exposure (under 5%), I would begin nibbling on the best in breed miners and or gold.

0 comments:

Publicar un comentario