RSI - Really Stupid Investors..?

Yep, after a month cycling in the Pyrenees, plodding through Sartre's finest works, and smoking Gitanes like they're going out of fashion, JdA is back at her desk, and is rather agog at the sudden outbreak of morale in the silverogosphere.

Indeed, thanks to some decent rises in the price of silver and gold during August and early September, we're all suddenly being advised to jump back in, with a range of erstwhile blog hosts queuing up to tell us that the 'doldrums' are over, and that the bullion banks have 'lost control'.

Silver in particular has had an impressive rally, moving from around $26.50 to $32.72 today. That's a whopping 19%. Well done to all who bought in at the bottom, and who are scaling out at the moment.

And a big wooden spoon to those doing their usual trick of beseeching their readers to buy as much as they can now that the rally is nearly at an end. You think I'm exaggerating? Read on...

The biggest problem for over-eager retail investors is that they have a tendency to buy high and sell low. In other words, they chase the momentum (buying high), and then bail out in a panic when things look rough (selling low). Despite being fully aware of this phenomenon, I've been known to do the same myself, so I'm casting no aspersions on the retail investor here. However, there are some tools of the trade that can help one to avoid this kind of behaviour.

The 'Mayday Massacre'

One of my better trading moves was bailing on silver in the high 40s during April 2011, and this was prompted by two indicators. The first, which I think I've mentioned before, was the 'credit card test'. I have a rule which helps me to keep emotions in check when it looks like I'm onto a winner: the first time that I see a reference in an on-line chat room, or on a blog, or a stock board, or whatever, which says that the poster is going to max out his/her credit card to buy the stock/commodity in question, well, then that is the very moment that I log into my trading account to dump the lot. The credit-card poster is an internet-age version of the shoe-shine boy, I suppose, but depressingly more naive and ill-fated.

The second, and altogether more scientific, is the RSI, which - in loose terms - helps one to assess how far above or below the trendline a particular stock/commodity is at that moment in time. Basically, when a stock chart starts having to 'shade in' its RSI, then you know it's a pretty fair bet that there'll be a major correction coming up (if it's above 70 or so) or that it's a decent time to buy (if it's below 30 or so). In April 2011, silver's RSI went off the scale:

Sadly, no commentator on the silverogosphere decided to address the RSI 'problem' in April 2011, and instead blindly encouraged their readers to go 'all in'. Memorably, Turd Ferguson advised his readers to buy a double-leveraged silver ETF (AGQ) virtually on the eve of silver's collapse. Harvey was convinced that the end of JPM was nigh. And Silvergoldsilver could bearly [sic.] contain his glee. Would that just one of the whole sorry bunch - self-proclaimed experts in the precious metals markets, let us not forget - had noted the glaring, 200 dB red-and-blue-flashing signal of the RSI. One glorious exception was Adam Hamilton at Zeal, whom I recall being widely mocked for daring to suggest that silver was heading for a serious fall.

The 'Brazilian Independence Day Slump'..?

Fast forward a year and a bit, and we're in September 2012. Let's have a gander at the chart...

As is obvious, the silver RSI is screaming once again. And this is no surprise: look how far it has moved north of its 50-dma. This is a tinderbox waiting for a match.

The tinderbox analogy is apposite. The RSI by itself does not guarantee a crash any more than it precludes continued massive gains. It is, after all, generated by a simple equation involving moving averages. Events, sentiment and fundamentals can all over-ride a simple bit of maths. So let's continue in the vein of a torturous Turd Ferguson-style analogy:

Imagine that your investment is a bale of hay. Now, for some perverted reason, your hay becomes more valuable if it smells of petrol. The buyers all just love petrol-soaked hay. The more petrol-y it is, the better. So, every now and then you go and tip a bit of petrol on your investment when you have some petrol to spare. Then a bit more. And a bit more. Every time you do so, it becomes more 'valuable'. However, and sadly, there is a mad arsonist living next door to you. He loves nothing more than playing with matches. You know that one day he will come and set fire to your hay (presuming you haven't sold it), but you have no idea what day that will be.

So, your quandary is, 'when do I sell my hay'? At what point does my fear of the arsonist outweigh my desire to keep adding petrol to my hay to make it more valuable?

The RSI is basically our indicator that tells us when the hay has become super-flammable. It is not saying that no more petrol can be added, nor is it saying that the arsonist is on his way. But what it is saying is that the moment that the match gets lit - and it will get lit, let's be in no doubt about it - then the whole farm is gone. Your petrol-soaked hay has now become too hot (metaphorically and perhaps literally) to handle.

Back to the real world. What could the match be? If it's the Friday jobs numbers in the US, then we could quickly find that that fills nicely the role of a match and we have a 'Brazilian Independence Day Slump' on our hands. Or maybe the match will come in a couple of weeks, with a pre-election effort to suppress the price of oil again. Or another fit of the Euro nerves. Who knows what the match will be? But in a world full of mad financial arsonists, who wants to be sitting on a hot commodity like the Devil's Metal with an RSI almost off the scale?


I'm writing this because no-one else is. It's no secret that I'm not a fan of silver, but I've no agenda here. What is troubling me is that although the charts are giving out the same signals once again, and that anyone who knows anything about the markets is just waiting for the match to come, the same culprits who drove their sheep over the cliff in April 2011 (whilst ironically calling those who refused to jump, 'sheep') are still trotting out the same perma-bullshit without a shred of responsibility to their readers.

I'm often criticised for goading the silverogosphere hosts, and sometimes I do stop and wonder whether I go a bit far. And then, in saner moments, such as today when I pulled up the charts from April 2011 and September 2012, and realised that not one popular silverogosphere site is seriously warning its readers of what might happen very, very soon, I think that I've failed in not going far enough.

Screwtape didn't ask to guard the guards, nor do we have a mandate to do so. But someone has to hold the silverogosphere to account for its irresponsibility and its ignorance, and to take the hysteria out of the advice peddled to the retail investor, n'est pas?


Warren James said...

Hey JdA, welcome back :)

Great analogy, would have to agree, and anyone out there thinking of ignoring these kind of technical indicators needs to consider that the large funds moving the price will most likely incorporate the RSI analysis and will sell or short whenever it suits them to, in bulk (despite the euphoria). It is like what milamber has said on the other post just now: "passion doesn't trump logic or change basic facts". And in today's world, the wave of sentiment is merely an artefact and is surely managed and monitored for best return (using advanced software, social media manipulation, etc).

But also: the 30-year cup and handle is just about perfectly formed and is shooting up around about now. I wouldn't mind betting that we see some really insane extremes before the dust settles on this one. I reckon we go over $50 before the year is out. It will at least redeem me for buying $40 silver for my in-laws fund ;p

Don Smith said...

To Turd's credit this time around, he has just posted a warning that we could see a drop in the next week.

The problem with all of these technical indicators is that when you go back and try to figure out when the best time to bail or buy, knowing the future, you still won't get it right. If you are just now buying, then obviously, it's best to wait for the inevitable pullback (normal market correction or "cartel smackdown" is irrelevant), but if you're buying for the long haul, it shouldn't much matter. We're as likely to, as Warren points out, miss a surge based on a multi-decadal pattern as to catch a new short-term bottom.

I agree that the bugs are too wrapped up in the momentum of it all, perhaps selling their book, perhaps true believers.

I for one watched the ride up, waited for the peak, watched the crash and, like Warren, bought some comfortably at $40 when it looked like the rise was about to resume. D'oh!

But I've bought some now in the 20s, too, so I'm feeling better. Still waiting for $38 to hit my cost basis, I think...

Anonymous said...


Thanks for the comment. I know what you're saying, but here is what Turd said today about the likely future direction of the PM markets:

It's great to see gold back above 1700 and silver is pushing up against the magical 33. These levels could very easily be the basis for a consolidation phase and, if the BLSBS comes in stronger than expected, they might very well be. But the charts look terrific and, as stated in the previous post, the CoT likely has room in it for a further extension of this rally before The Cartels aggressively step in. Therefore, I still think that gold may see 1750+ before any type of significant raid occurs. Silver could easily extend toward 35, too.

Hardly a serious warning, is it? And certainly not a discussion of a potential 13% (or more) drop in silver... And, crucially, absolutely no mention of the RSI. Turd loves the RSI when it gives out a 'buy' signal, but apparently thinks it's irrelevant when it gives out a 'run for the hills' signal...

Anyway, as always, I'm just trying to write something for readers to chew on.

I hope you get your money back one day. I'm sure you will, for what it's worth. It could be in a month, or it could be in three years. But you'll probably be ok.


Jeff said...

The fed won't act any time soon, this close to election, with equities making new highs and oil/commodities far higher than when they last acted. The only thing propping this market has been hope and the ECB. If employment comes in high tomorrow that will prove even more that the fed wont act. If numbers come in weak ie less than 80, then Id consider going long, but Id give myself lots of time ie march 2013 contracts. If numbers come in high and the rally someone continues then I will fade it in a major way. I've already started lightening my metal positions.

For the record, does anyone actually consider turd (a regular guy with a macbook ) or sgs (a washed up pissed off trader) experts on the subject? I surely hope not.

Kid Dynamite said...

A few words on RSI, if I may:

I would recommend checking out JC Parets at on the subject. What he looks for is when RSI direction diverges from price. ie:

just looking at your charts, that doesn't seem to be really happening yet in silver.

What I see as the risk to silver is not RSI, but rather: UBI: the utter bullshit index.

The amount of "flat out making shit up" in the silver ogosphere is reaching a crescendo... MS is about to go bankrupt? JPM is "losing control" of their "short" position which, by now it should be obvious to anyone with a clue, they don't even have (no huge net short, I mean, despite the short COMEX contracts)...unnamed buyers aren't getting delivery? it's taking Sprott forever to receive PSLV's silver (false: Sprott hasn't bought all the silver yet! and he said it took two weeks to get the stuff he bought...)

or today, Egon Von Greyerz just flat out MAKING SHIT UP:

""Today Egon von Greyerz told King World News, “... we are now seeing a lot of fund managers and investors moving out of gold ETF’s, and taking delivery of physical gold and holding it outside of the banking system.”"

flat out lie - just look at GLD inventories SKYROCKETING lately. the OPPOSITE of investors redeeming the gold ETFs...

that's what worries me...

it's like early 2011 all over again (not sure we're in April, yet...)


Bron Suchecki said...

Good post, blogged on it here to help get the message out

S Roche said...

Hmm,timing is everything.

The 4hr chart had worked off its overbought RSI, and while you are at it take a look at the Spot Silver monthly RSI...

Anonymous said...

How's that top calling working for ya?

Anonymous said...

Nice picture, Happft :-)

I think you need to re-read the article. It wasn't a top call. I was simply pointing out that it is a dangerous time to be adding to a position in silver, given that it has moved nearly 20% in a few weeks and the fact that the RSI is screaming 'overbought'. The move of today simply makes it even more 'overbought', and anyone chasing the move is, in my opinion, playing with fire (or petrol-soaked hay, if you like). And anyone advising readers to do this is - in my view - being irresponsible, just as they were in April 2011.

I also made it pretty clear in the article that the timing of when the match will arrive is impossible to predict. What the RSI is telling us is that when the match arrives, it's going to be bloody. Silver now has even further to fall, thanks to today's 'headshot'.

I just took a look at how silver's RSI is looking following today's move, and it's scary. Also worrying is the fact that gold has now moved into overbought territory too.



Anonymous said...

@S Roche: Fair point. It's always the trickiest aspect of chart analysis - deciding which timescale to use. However, it is really just a matter of how long-term one is looking. For example, if one uses the daily (as I did), then it puts the timescale of both the big move up and the (potential) move down onto a period of weeks.

So, yes, using the monthly, one can imagine not much of a cumulative move over years. But the point is that blog hosts are advising people to buy NOW. Why do that, if a correction seems likely?

It is the attitude of 'don't miss the train' that makes (a) bubbles, and (b) broken retail investors.

Don Smith said...

@Kid Dynamite - Couldn't agree more. I don't take anyone on any blog at face value when it comes to market timing. It's a fool's errand and a dangerous one at that.

I don't buy the Cartel hype, either. There are clearly aspects that are true, since the Blythe Masters assertions that they only hedge is patently false, and the price action is clearly manipulated, or at least news is being front-run.

However, Silver at $100? The dollar crashing? Maybe? I don't know. I buy PMs to maintain a zero-counterparty risk investment that will hold value, not because I think it will go all tulip-bulb on us, like a year and a half ago. I thought a 20% smackdown was a decent floor, and I certainly think $26 is the long-term bottom. But I am sure there will be opportunities to stack around or below $30 again. Maybe next week.

Anonymous said...

Just trollin' with you, your humble attitude however rendered my effort mute.

Byzantium said...

Hi JdA,

as a self confessed PM bull, I do distinguish between the short, medium and long terms, and I value Screwtape for its contrarian balancing effect.

When confronted with price strength as we currently have, I try to allocate my tradable portion to feed a stream of sell orders, in a bell curve formation, with the height of the bell positioned where I anticipate the top will be. Of course, I am crap at calling those, but am currently targeting $40 and $1800.

RSI is certainly one to watch; thanks for the tip.

zoggl said...

There is an argument for everything. Look at the RSI of silver in 2010 on the daily chart - a few times overbought from 20 USD to 49 USD. Retracements where shallow. If overbought conditions are overrun to the upside its also a sign of a very strong trending market. Who knows? It's a game. I see a break out from a long consolidation and hope I can survive the dips. Remember: Its a bull market, so it doesn't pay out to trade in and out to much. My opinion: This is all inflation driven, and the real inflation is not priced in in the PMs.

Anonymous said...

@Byzantium: A very sensible-sounding strategy. Your targets might be a bit on the optimistic side, but my crystal ball is, of course, no clearer than yours. Regardless, scaling out as the RSI gets especially high (and scaling in when it's especially low) is a strategy which has worked consistently well for me over the years. Most importantly, it's really saved my backside a few times..! (Silver being a case in point, but there are others, such as LON:BOR - see here).

@zoggl: we'll have to agree to disagree on the inflation point. Let's take gold as an example: if it's risen by 20% a year for the last ten years, then it has remarkably exceeded the rate of inflation, regardless of which measure of inflation you choose to trust. If your investment thesis is that the gold price is supported largely by inflation, then the gold price is on extremely shaky foundations at the moment. It should be somewhere between $500 and $1000 an ounce..!

Anonymous said...

I've just posted a bit more analysis on the silver price at Screwtape's new repository for short-term trading information, JdA's Trading Set Ups, here.

daffy said...

RSI always worked better as a buy signal for me than a sell. I do not trade anymore since the thefts of accounts, now made legal from what I read. Never visit Turd anymore but keep up with most of them otherwise. Really like your site. Most of my physical was bought at $5, before the Y2K hype and I sell just a few ounces at a time here and there as 40 FRi night for $35 oz. Keep up the good work.

Funky Tape said...

Victor - how exactly do you check the GLD inventories? Been waiting for you to update your most recent post from June with the sweet graph in it, but it seems to be stuck in a time warp. Thanks.

Funny, I like to pile about 5-6 momentum indicators into TOS at once but leave out RSI on purpose as I've never found it useful.

One thing you can bet on with about 99% accuracy is consecutive daily closes outside the bollinger band. The most I've ever seen on any chart with anything is 7. Silver just did 6 and also had a 6-run from $40 to $49. I doesn't mean the run is over, but some type of sideways moment is essentially guaranteed, albeit maybe for less than a day.

And what, no mention of Bill Murphy on Capital Account recently stating that silver would be $60 "in a few weeks"?

Anonymous said...


I am afraid I won't regularly update that GLD post. People in the comment section at FOFOA's, however, often post new signals (including me). But as nobody would trade that strategy anyway, this is just for fun.

If you take a look at my post, there is a link to an Excel spreadsheet on the GLD website that has all the information that the strategy requires. Yes, all this was done with a single Excel file.


Don Smith said...

Today is why technical indicators are crap. Of course, touters of one investment or another (PMs, stocks, bonds, timeshares) shouldn't point to charts, either, but this market is broken.

Anonymous said...

@Don: au contraire. Nothing happened on Thursday/Friday which broke anything. I notice that silver, many (if not most) of the miners, and gold, all stopped at highly predictable fib lines and other resistance lines. The perfection of that move was astounding, in fact. I hope to put a post out on it later this weekend.

As I try to explain in an update at my spin-off trading set ups site (JDATSU), I was looking out for a potential short on silver when it gets to $35. Well, Bernanke or no Bernanke, we've still not even got to $35, and we may (may) already be on the way back down. So I'm waiting with interest to see if there's enough fuel in silver's tank to push it through that line. With the indicators being what they are, I suspect not...

In short: nothing is behaving in any way unusually in terms of fibs or indicators.

S Roche said...

"still think that gold may see 1750+ before any type of significant raid occurs. Silver could easily extend toward 35, too."

Quite a polished performance, in hindsight!

Dan D. said...

Great piece as usual. I have been less active in blogging at the moment but now that we have dodged a family crisis I look to get back into it in full swing.

Always great reading your information and well informative pieces my friend.