The Big Dipper

What does the well known constellation we all learned to look for in the night sky have to do with the gold market? Answer - Nothing except it sounded like a catchy title for a short missive on the recent doings in the gold market!

On Tuesday, February 7, 2006, gold experienced the biggest single day decline in price in dollar terms since August 1993. From a closing price of $574.30 basis April the previous day, gold closed down $19.50 at $554.80. This price drop came on the heels of the yellow metal making a new 25 year high at $579.50, a mere 3 trading sessions earlier.

The drop has rattled even some stalwart gold bulls while not surprisingly bringing gold bears back out of hibernation once again with their dire predictions of a massive, deep correction lying ahead. I have seen some of the usual suspects calling for price drops of as much as another $100/ounce from current levels. Some of the perma gold bears are even chattering about $375 gold again.

Of course many of these people love to give their caveats and qualifiers by stating that they nonetheless remain long term bullish on gold and fully expect a resumption of the long term bull trend at some point in the future. Comforting words indeed!

What I would like to demonstrate in this missive is that far from heralding the demise of gold, violent downward reactions of this nature are going to become more and more the norm as gold firmly establishes itself in this phase of its generational bull market.

The reason for this, in my opinion, is that we seem to have entered an era in which extremes in both price action and volatility can be attributed directly to the plethora of computerized trading platforms and system traders which are increasingly coming to dominate the markets.

Try to imagine if you can thousands of computers running technical analysis software which are tied directly to online trading platforms. These electronic trading systems are designed for speed and objectivity. In other words, they are designed to eliminate any guesswork and/or human factors when it comes to the actual execution of trades.

The way these things work is pretty straightforward. A hedge fund for example will purchase a system, whether for intraday use or longer time frames, which is programmed with a set of parameters that it tracks in order to generate buy or sell signals. These signals are all based on price movement of some sort. Some of the systems will trigger a buy signal for instance, if the price exceeds the high of the last 5 days or a sell signal if the price drops below the low of the previous 5 days. Others are keyed to moving average crossover patterns. Still others are yet keyed to oscillator crossovers. The point is that such systems can easily and quickly be set up to automatically scan price action in any given market that is desired to trade in for such signals.

Now comes the interesting part. These trading systems have the capability of interfacing with the trader's electronic trading platform so that they can AUTOMATICALLY send the buy or sell signal directly to the exchange without the trader even having to manually enter the trade himself. The computer automates the entire process. The goal in employing such technology is to gain an advantage over the next guy by beating him to the punch and getting your order to the exchange first and as quickly as possible. The name of the game is speed and if you have to wait around for some carbon-based human life form to actually key in the order and send it, you have forfeited another 10 seconds or more - valuable time in which thousands if not tens of thousands of dollars can be lost due to execution lags as price begins to move leaving you a much poorer entry point.

Realistically speaking, this technology allows the hedge fund manager to be out on the golf course having a drink at the club house while his computer is sending orders to the exchange on behalf of his clients! Not that they are doing that (although the terrifying thought is that there probably are some that do!) but the potential is there. After all, once the system is set up properly, the entire system is fully automated.

At exchanges where the contracts are traded upon a fully electronic platform where orders can flow directly into the exchange's order system, the result is a near instantaneous execution of an order that had a buy or sell signal generated only mere seconds earlier! You want to talk about lightning fast!

Think about this for a minute and you will begin to understand the repercussions of this development and how it can result in unprecedented barrages of orders flowing into the exchange at any given moment in time. In time past, orders had to be phoned in to a broker's desk who then phoned it down to the floor which then sent the order into the pit via a runner to the floor broker who then executed it. The process had by today's standard, a huge time lag. Not any more. The process is frighteningly efficient.

Now, I submit that it is this very process which lies behind the unbelievable wild swings and huge intraday price movements that mark so many of today's markets, especially the futures markets. At any moment in time, thousands of these computers are tracking the very same price data in a given market. Any move that therefore triggers a sell signal on one will soon trigger a sell signal on another. That will trigger yet another sell signal on a third and so on and so on and so on.

In reality the sell signals are being triggered on more computers than one at a time but you can understand my point here. With all these fully automated trading systems sending their orders to the exchange within seconds of one another, all price movements become incredibly exaggerated as sell order upon sell order is piled upon the next. The net result is that existing bids can be swamped in a matter of mere seconds producing a snowball effect that quickly becomes an avalanche.

This is precisely what we witnessed in gold on Tuesday, February 8, 2006, this week. It is also the exact same thing which we have seen for better than a year in the copper market as well.

The interesting thing about this is that while the intraday sell offs can quickly cascade out of control, the price reactions seem to be finishing up much swifter than they have done in the past with the result that the primary trend is reasserting itself much sooner than was the case formerly. In other words, though the price reactions tend to be more brutal on an intraday basis, they also are completed in a shorter time frame before the market resumes the direction of the previous trend.

What I am advancing here is the notion that since so much of human judgment has been taken out of the picture due to this automation, there is little need for decisions by trading funds managers the following day or even the days after the price reaction. The computers take care of the whole kit' n' caboodle. Whereas formerly it took fund managers some time to evaluate a market and decide whether to sit tight or lighten up, that is no longer the case. The entire process has been taken out of their hands and given to a computerized platform to perform. It is done- caput, finished - objectively, swiftly and efficiently. So much so that I would venture to say that in many cases, hedge fund managers have been reduced to being mere spectators whose primary role is to pick the markets they want to trade in while leaving the actual buy and sell orders to the programmed trading system they are employing.

We then as a result get the swift, sharp downward corrections that we now have come to know only to witness the market within a matter of days go right back to trending as if nothing whatsoever had transpired. All the selling is taken care of at one time without any attention paid to subtlety or finesse. It simply gets done as the computer is not programmed to get its job done with skill - it is incapable of that - it only knows to sell and so that is what it does and it does that automatically by continuing to send its orders to the exchange until it has exhausted them. Unless that same system then somehow gets flipped to generating a new signal to go short, the selling is exhausted within a matter of days and there is no one left to further sell the market. At that point, fundamentally oriented traders looking to buy jump back in and the market then reverses itself and heads back up with the result that the same hedge fund computerized trading systems are now sending buy orders to the very market that they just finished selling a few sessions ago. Traders attempting to short these markets are literally blown out of them and forced to cover much to their stunned amazement. It is quite a strange sight to me personally to have observed this phenomenon which I first thought was a bit of an anomaly but am now coming around to believing is becoming the new norm for many of today's markets.

Let me give you a charted demonstration of what I am stating here by using another market that has been in a primary bull trend for some time now and that is the copper market.

The entire upward course of copper over the past couple of years has been punctuated by these days in which we have seen huge downdrafts only for the market to apparently shrug them off within days and then proceed to make new high after new high.

Take a look at the charts below and you will see these some of these big down days noted for you.

Notice that I have included the total amount of money involved per contract to give you a sense of the magnitude of the move in dollar terms. In each instance you can see that the one day or combination of several days resulted in huge paper losses for all the longs. To get a perspective, the single day drop in early October 2004 in the price of March 2005 copper, some $15.80 resulted in a paper loss of $3950 per contract for every long. A loss of that magnitude would have required at $39.50/ounce drop in the price of gold in just one day to match that kind of paper loss.

We see something similar once again in that same March 2005 copper contract in early January 2005 which gave us a $15.00 down draft in one day or a paper loss of $3750 per contract. Such a loss would have required a $37.50 down day in the price of gold to equal that size loss.

Note also the big downdraft in December 2004 which took place over a period of 5 days. Again, sizeable losses if you were long. But in each and every case, notice how swiftly the price of copper recovered to go on and make new highs. Imagine the chagrin and absolute frustration which greeted any would-be copper short who attempted to enter the copper market the very next day after the big down day.

Now take a look at the September 2005 copper contract.

Note the sharp sell offs and the resulting paper losses for longs during these periods. Yet note how quickly the market recovered and once again went on to make new highs in a relatively short period of time after what appeared would be a debacle for the longs. Opportunistic shorts who tried their luck at this market soon learned the hard way what it means to try trading against the primary trend. The big two-day downside cascade of June 30 and July 1 was completely erased in the next 4 trading sessions. It took 9 days to undo the $8.20 swoon in mid August but nevertheless, it too came and went.

Now look at a more current contract - March 2006 copper - and note the months of October and November of last year (2005). October's two day swing resulted in a paper loss of $2812.50 per contract for the longs while November's 3 day swing was good for paper losses of $2612.50 per contract. Again to put things in perspective by comparing this to gold - losses of this size would have required a $28 downdraft in two days and a $26 downdraft in three days in gold to equal the size of such paper losses.

Folks this all seems quite cold and distant when we can sit here and objectively analyze these things but I can assure you, any trader who is long and is stomaching those kind of paper losses is anything but calm and serene while this is occurring. It is absolutely guy wrenching to witness these kinds of swings in one's trading account. An element of distrust is always introduced; "Could I be wrong? Is the move up finished? Is the bull market history? Where are all the buyers?" Those are the thoughts that a trader grapples with after a brutal sell off of this sort. Yet the amazing thing is that in both cases in this contract, the losses evaporated within days and the market stormed on to make yet more new highs. The ferocity and intensity of the sell off was quite severe while it was occurring but it was also incredibly short lived.

Look at the last ellipse which is about as current as we can get. Copper was not left unaffected in the rout that hit the commodity world on Tuesday. It was tagged for a swing of some $10.10 in the last two days or $2525 per contract. Again that translates to a $25 sell off in gold to equal such a paper loss. Will copper move on up once again making this latest sell off just one more in a line of brutally stunning yet short lived price reactions in what is a raging bull market? Time will tell but the odds sure favor it to do so.

The point I have been attempting to make here is that the computerized trading systems that no doubt are being employed by the macro funds playing copper have sent many a sell order at the same time on the same day into the copper pit over the last year. The result has been stunningly severe sell-offs which rocked the bulls' convictions to the core but nonetheless, when looking back, were nothing but momentary blips on the radar screen of this massive bull market that copper is experiencing. Each correction instead of lasting for weeks on end has been relatively short-lived with the primary trend quickly reasserting itself within a short period of time.

Looking at this particular market it certainly seems that my observation about price reactions being brutal yet short lived has some credence to it. But is this phenomenon peculiar only to copper or is it perhaps the beginning of a more common pattern of market behavior? Are these computerized systems so efficient at their tasks that they can actually condense heretofore reactions into a shorter time frame?

I am not yet sure about this to be perfectly honest but I do find that many different commodity markets seem to displaying this same sort of pattern of late - Strong, sustained up moves accompanied by swift, sharp and brutal sell offs which run their course in the matter of a couple of days or a week at best before the bullish uptrend reasserts itself.

I strongly suspect that gold is not going to be immune from this pattern. Why? Because all of these various commodities that exhibit this pattern have one thing in common - the fundamentals behind the upward move in price are so strongly bullish that speculators, particularly hedge funds, want in no matter what. Demand seems insatiable and rising prices are simply not slowing it down. Simply put, people want the stuff and keep paying up to get it. While the downdrafts are gut-wrenching, the setbacks in price are viewed as buying opportunities by those either wishing to get in or those wishing to add to their existing long positions. The computer generated sell-offs flush massive amounts of positions out of the market but demand is so strong that fundamentally oriented buyers rush back in on the heels of these sell-offs. Their buying then flips the trading systems back to the buy mode resulting in an avalanche of new automated buy orders. These in turn drive the price on up to new highs where the advance temporarily stalls out as the market pauses. Then the entire process repeats itself all over again as the sell signals get generated and down it goes again only to pop back up like a cork in the matter of a few days. It is really something to behold.

The conclusion to this is simple - get used to this wild volatility in the gold market - it is here to stay and is only going to become worse the deeper we get into this phase of the bull market and the more that the bullish fundamentals intensify. Be very careful about heeding the siren songs of those who are willing to call for a top in gold every time it experiences a significant down day or two - remember copper! The more macro funds that are attracted to gold, the more open interest is going to surge. The more open interest surges, the more volume is going to increase. The more volume increases the wider the daily trading range is going to become. The wider the trading range becomes, the more chance there will be for automated trading systems to generate buy and sell signals. The result - more volatility and wicked price swings on both an intraday and day to day basis.

Trading discipline is essential if you are going to survive this period and prosper. That means respecting trendlines and watching for support and resistance levels to develop. Watch for stalling markets in particular as those will be the time periods in which the computerized trading systems are the most dangerous since patience is not one of their virtues.

Novices and even some of your more experienced traders would do well to keep this in mind and be careful about the position size you put on. We are entering the period in which gut-wrenching is going to be commonplace and unless you want to live on a diet of Maalox or Alka-Seltzer, it will be prudent to know your emotional limitations.