The COT charts faithfully track the ebb and flow of what might be termed "smart" and "dumb" money in the market, and thus give a strong indication of when a top or bottom is forming in a market. On these charts the market players are classified either as "Large Specs", "Small Specs" or "Commercials" and it is very easy to label these groups of players as either smart or dumb by empirical observation - by going back and seeing the positions they took in the past and what subsequently happened. It may appear to be cynical or even cruel to label market participants in such a seemingly cavalier fashion, but the market is a cynical place, and anyway it's fun, and it's not personal - the composition of these groups is continually changing and a trader may begin his career as a dumb small speculator and slowly graduate to become a commercial and spend his later years fleecing dumb small (and large) specs.

From the foregoing you will have already gathered how the groups are classified. As we will see it is plainly obvious that both the small and large specs should be classified as dumb, and the commercials as smart. Just to add a bit of color, we will label the small specs as "sheep", the large specs as "oxen" (you've all heard the saying "as dumb as an ox") and the commercials as "wolves".

So, without further ado, let's get on and see how the sheep, oxen and wolves have fared in the gold market, as revealed by the COT chart. As we can see on the accompanying COT chart, the sheep are permanently net long, which is pretty dumb when you stop to think about it, and compound their stupidity by having the largest long positions at market peaks. So our first useful observation is that when their long positions are at a relatively low level, the market is likely to be near a bottom. A more dramatic indicator of the state of the market is afforded by the oxen, however. The net long positions of the oxen build up to a huge level as the market approaches a peak, and then ebb away to a low level as the market approaches a bottom. The chief difference between between the sheep and the oxen is that the sheep are just plain incompetant, whereas the oxen are "professional" incompetants, who are distinguished by making losses on a grand scale, which is usually acceptable because they are doing it in company and with other peoples' money. The wolves (commercials) are statistically always on the right side of the trade and systematically relieve the other groups of their funds.

Now lets look at a couple of examples of what I'm talking about on the gold COT chart. This COT chart goes back to 25th May 04, and a chart for gold that goes back to about the same time has been placed beneath it for direct comparison. On the chart for gold itself we see that gold bottomed at about $370 (right on its long-term uptrend line) early in May, and subsequently rallied right through the summer and fall, peaking at the end of November at a new high at about $460. Looking at the same period on the COT chart we see that the oxen had a relatively very low net long position at the market bottom, and a comparatively huge net long position as the market peaked. The wolves, on the other hand, had a relatively low net short position at the start of this period, and a comparatively large net short position at the end of it. You can't ask for a much more dramatic example of this of who makes money and who loses it in this market.

Our 2nd example is provided by the period from early February through early May this year. As gold declined during December and January, the oxen greatly reduced their long positions and the wolves covered their shorts, which dwindled to a low level just before gold broke strongly higher in February. As gold rose sharply into March, the oxen once again balloned their long positions as another peak approached, while the wolves ramped up their short positions. But what is especially interesting is how the long and short positions of the respective groups continued to climb into early May, because the symmetrical triangle formation in gold was due to be resolved with a breakout - and break out it did, to the downside, to the dismay of the oxen who turned out to be on the wrong side of the equation, as usual, and to the glee of the wolves.

It should now be clear to readers that this is a very reliable and practical way of determining what is likely to happen in this market, and in other markets where these statistics are maintained, but before concluding by extrapolating the outlook for the market, which is the ultimate object of the exercise, there is one more useful indicator on the COT chart that we should take a look at, and that is the "net open interest" line, which is the wiggly green line on the chart.

The net open interest line is very easy to interpret. The scale for this line is on the right side of the chart with the green numbers. This line is at a low level at market troughs, for at market troughs the sheep and oxen hold a small number of long positions, and the wolves hold a reduced number of short positions. At market peaks the sheep and oxen, especially the latter, hold a high level of long positions, and the wolves hold a large number of short positions, consequently the line is at a high level, as was the case when it peaked last November. Therefore we have good evidence of a market bottom when this line is at a low level.

Now to extrapolate the outlook. We have already observed on two occasions that when the oxen hold a very low level of long positions, and the wolves have a very low level of short positions, the market is ripe for a significant advance. These occasions were May last year and early February this year. Note, however, that net open interest was nowhere near as low in February, which preceded a sharp bounce, as it was in May last year, which preceded a major uptrend. On the COT chart we can see the long positions held by the oxen and the short positions held by the wolves both declining in a very satisfactory manner, however, there is clearly scope for considerable further decline, implying that we are likely to see some further decline by gold in coming weeks, which is hardly surprising considering the implications of the breakdown from the symmetrical triangle. The open interest line, which has not fallen that much in recent weeks, also has leeway to drop considerably before a bottom is in. Gold stocks have massive underlying support (around 150 on the HUI index) and so may not suffer unduly if gold drops back to the $400 area in coming weeks, which would be expected to throw up a major buying opportunity (with relatively close stops) in the stocks.

In conclusion the COT chart suggests that a bottom in gold is close, but that we are not quite there yet. Stay tuned.