Commitments of traders. Useful or useless? I've seen arguments on both sides. I think you have to take the data and how it can be used and examine it on a market by market basis. This is definitely not a one size fits all situation. The common simplistic approach to COT numbers is to follow the commercials, fade the small specs, and the large specs are sometimes a fly in the analysis ointment. But this approach fails miserably in some markets, and the question is: Why? The 1st question you have to ask when analyzing the COT numbers for a particular market is "Is this one of the markets where primary price discovery is being made for this particular commodity?" If the answer is "No." as in the currency markets, where the forex markets dwarf the amount of volume in the futures pits, you have to throw out the idea of following the commercials. Their open interest in the futures market could represent just about any form of transaction and is just a tiny blip on their balance sheet. So you're left with the retail and non-commercial numbers. Non-commercial numbers could easily be an arbitrage of some sort, so that doesn't do you much good, and the retail traders catch currency trends and ride them to a profitable outcome too often to be a reliable fade in my opinion. So the answer for currencies would be that the COT's really aren't of much help. The next question is "How do the retail traders react to this market?" The reason I bring this up is that there are quite a few markets that tend to catch the eye of the crowd and knowledgeable or not, the crowd can overpower the strong hands just through force of numbers. This happens in highly visible markets such as gold, and low margin markets such as oats and rice are particularly susceptible to this phenomenon. Also, oddly, bean oil is one of these markets. To the rest of the grain world, bean oil is a byproduct, but retail traders really like it. I suppose because most people have a bottle of it in the pantry. Retail fever has blindsided many a pro, myself included. (The great Japanese rice crop failure of '94. The Japanese don't even eat our rice for Pete's sake!) So, standard COT analysis can even be dangerous in some markets. The last thing you have to ask is "Is this market heavily dominated by one group?". This would be the case in natural gas where the commercial trade frightens all but the foolhardy out of the market. Is COT data useful? Sure. Show me a heavy one sided commercial position against an opposite spec position in a wheat, cattle, sugar, cotton, etc. and I'll happily take the commercial side. And watching the net position changes in the markets that are amenable is very telling data. You just have to keep in mind that it's a hammer, and not all the world is a nail. Matthew Shelley (broker) Futures and options trading involves substantial risk. Information is obtained from sources believed to be reliable and is in no way guaranteed.
I don;t have much knowledge of the effectiveness of the COT for the grains, energy or interest rate markets, but I usually will take a look once a week for the stock indicies. I don't really know if there is any real application for the numbers...I figure that a large majority of any small speculator open interest in the ES could either be from member arbing against the big contract or some other type of hedge that would not be just a pure directional play...As time goes by, the markets become more and more sophisticated as far as using different instruments for altogether different means...Maybe 20 or 30 years ago before all of the hedging, arbing, etc...the numbers offered a "pure" perspective or insight into the trading bias...
I don´t trade commodities, but after that caveat, I believe that only extremes of COT are useful as signals.
matthew, I don't have your experience in the ags and softs but I attempted to follow COT data for about a year using a highly regarded newsletter and found it no better than a coin toss. It's analogous to insider trading data for stocks. When it works, it seems totally obvious but there are plenty of times when it hides more than it reveals or what it reveals is already in the market.