Pinpointing Participants

Analyzing Futures Markets The Upperman Way

By Floyd W. Upperman JR., CTA

“We are participants, whether we would or not, in the life of the world…”

--Woodrow Wilson

Police detectives are taught that if you want to know who’s behind a conspiracy, the first thing you do is “follow the money.” In the futures markets everyone seems to be following the money by analyzing the price movement of individual markets. But if the answer to profits in the futures market were simply a matter of price analysis, why doesn’t everybody get rich by just betting on the direction of the price trend? It’s because simply understanding price, i.e., following the money, is only part of the equation.

If I were teaching police detectives how to understand what’s responsible for the action in the futures markets, I would tell them to not only follow the money, but to follow the market participants. By tracking and studying the activity of individual market participants in a given futures market, a commodities trader can gain certain insights into each market that can be extremely helpful in forecasting significant trend reversals and future bull or bear markets. Fortunately, you don’t have to be a Sherlock Holmes to learn how to do this kind of analysis. All you need is a bit of raw data and the know how to interpret it.

Each week we commodities traders receive a gift from the U.S. government. No, it’s not a special tax break reserved only for registered CTAs. It’s the public release of valuable “inside information” identifying both long and short positions held by individual market participants. Uncle Sam has been providing this information to the public—free of charge I might add—in its current format since 1983, and in other formats prior to 1983. We call this gift the Commitment Of Traders (COT) report.

The COT report, which is compiled Tuesdays and released to the public every Friday, provides a thorough history of commercial and non-commercial activity in each futures market. By analyzing this data traders can go back in time and see how certain select market participants prepared or positioned themselves ahead of significant market turning points and/or in front of extensive bull and bear markets. It’s much the way a military historian would study the positioning of troops before a decisive battle. Determining how the participants behaved before a known past outcome can help us understand, and even predict, what may happen in the future.

Unfortunately, many traders (both amateur and professional) do not completely understand all the benefits of tracking and thoroughly analyzing the COT data. You see, this is not just simple price data. It’s not just “following the money.” The data is complex, and a proper study of the COT report requires knowledge of the markets and at the very least a working knowledge of statistics. Fortunately, I’ve already done all the heavy lifting for you. Through a proprietary automated computer program designed to analyze COT data, tracking the results of this information is as easy as logging on to my website. In fact, I look forward (as do my subscribers) to Fridays, not because it’s the end of the workweek, but because we not only get our COT gift box, we get it unwrapped for us automatically.

When employing the COT in any futures market analysis, consistency is of the up most importance, and in my opinion it is one of the keys to success in this business. By consistency I mean performing the same study week after week, using the same parameters in order to obtain important knowledge and understanding of the unique behavior by individual market participants. Because each market is unique, the same participant behavior in one market may mean something entirely different when it occurs in another market. This “pattern seeking” as I call it in various markets is essential to a human being, whether he is trying to find order in the cosmos or a simple pattern in the price of pork bellies. It’s the same search for a predictive pattern of behavior.

While the Commitment of Traders tracks the positions (longs and shorts) held by all market participants, my analysis further breaks down this data and applies proprietary statistical measurements and indicators to identify trading opportunities. We combine these indicators with proprietary price indicators and graphically present the results to you.

Breaking the COT down to its elements, we see that there are three key market participant categories:

1) “Com” -- large commercial positions (producers & consumer hedgers)

2) “NC” -- large non-commercial positions (commodity investment funds)

3) “SM”-- non-reportable positions (small speculators and small hedgers)

Each category represents a particular group of participants. The Com category represents the commercial hedgers. The commercials are widely considered the most important group. The commercial hedgers are composed of commercial producers and commercial consumers of a particular commodity. Overall, this group is the most knowledgeable in each market, because their very livelihood depends on getting it right with respect to future prices. These two groups of commercial participants have different reasons for being in the market; however, both share the same goal, which is to reduce their risk in the cash market. For the producer, this may mean locking in a particular price using futures contracts to reduce the risk of being forced to sell their “produce” at lower prices in the cash market. For example, a gold mining corporation is a producer of gold. They will sell short gold futures contracts to lock in a price for their gold and thus reduce or contain their exposure to the risk of falling cash prices in the future. A commercial consumer on the other hand, is concerned about the possibility of rising raw commodity prices. They will also use the futures markets to control or contain this risk. An example of a commercial consumer for gold would be a large jewelry maker. They need to be able to purchase gold to produce their jewelry. They may buy gold futures contracts to lock in the future price paid for their gold, and by doing so, reduce or contain their exposure to the risk of being forced to pay higher prices in a rising cash market.

By using statistical indicators to monitor the weekly changes in this data, and by comparing current conditions with historical conditions, one can identify statistically unusual activity as it is occurring. The main group to monitor is the commercial hedgers, as they tend to be the most knowledgeable. They are either producers or direct consumers of the raw commodity, and deal in the cash market regularly. The producers tend to know when production is down just as the consumers tend to be most privy to swings in their customer demand. The bottom line is they are in the best position to be able to foresee changes in demand and/or supply.

Unusual commercial activity is defined and measured using statistical formulas, and is often indicative of major tops or bottoms in a market. Let’s see how we can recognize these phenomena using examples in both the Nasdaq and cotton futures. First, we will look at a simple distribution of the commercial positions in the Nasdaq 100 (See fig. 1). In this graph, the average net commercial position is in the center of the distribution (bell curve) and the extreme conditions are found in the tails of the distribution. The more rare or extreme the position, the further out in the tails it is found. It is during these times of extreme commercial activity that major tops or bottoms have a tendency to form (See figure 2 below).

The large build up of commercial longs shown in the Nasdaq 100 (UCL/LCL) was suddenly and abruptly reversed by the commercials on or about April 4, 2000. This coincided with the top in the Nasdaq 100 futures. At the time, the public was excessively bullish the tech-heavy Nasdaq 100. Based on the above data combined with our price data, we were bearish at the right time. We were successful in exiting our long positions near the top, and we established follow-up short positions as prices began to descend.

Sometimes this pattern is very clear, as it was with the Nasdaq-100 and cotton late in 2000 and early 2001. The data clearly indicated a top in the Nasdaq 100 in April of 2000 and then a top in cotton prices in the fall of 2000.

The movement among the commercial institutions clearly indicated an imbalance was taking place in cotton futures late in 2000. This occurred long before the imbalance manifested itself into sharply lower cotton prices. This is precisely how imbalances are corrected. We observed a similar (yet opposite) paper imbalance in OJ in the late summer and fall of 2001 as well. However, OJ is not as liquid (no pun intended) as the Nasdaq 100, or even as liquid as cotton. Nonetheless, the imbalance in OJ was on the buy side and sure enough, in the fall of 2001 OJ prices began to rise sharply.

As you can see, by combining certain COT conditions with price behavior and price structure in a given market, we are able to identify what I have termed high-probability “setup conditions” based on similar past conditions that have produced predicted results. This is also known as an Individual Market Participant Activity, or IMPA setup.

An IMPA setup is a combination of technical factors based on price behavior and specific statistical requirements from the COT. If an IMPA setup meets our proprietary “criteria” it provides us with a high-probability “setup” condition.

During the early stage of an IMPA commercial imbalance, we don't know for sure if the paper imbalance will resolve itself without developing a new trend, or if the paper imbalance is something that will spill over into the physicals due to some real, and unknown, underlying issue being anticipated that is likely to impact supply and demand. In the case of cotton, unusual positioning of the commercial participants (broken down into their respective counter parts) indicated an imbalance in supply/demand was taking place (on paper) before prices dropped sharply—first in the futures and followed by the cash.

This is normally what we expect as the IMPA analysis provides more of a leading role and does not lag the price, as most price-derived indicators tend to do. The actual price structure and changes in price are important in determining the entry of a setup. One of the big advantages of including the IMPA with price derived indicators is that we are always paying attention to the market at the right time, and ignoring the markets’ technical signals when no real underlying supply and/or demand issues exist, thus bypassing technical smoke screens that are likely to fizzle or amount to little. In other words, not just following the money, but following the participants.

In cotton the imbalance noted in the futures markets via our IMPA analysis and documented in my reports clearly revealed that the imbalance occurring on paper late in 2000 and early in 2001 would result in grossly lower prices for cotton. This particular imbalance remained in affect for a fairly long period of time. Eventually what tends to happen over time is the initial reaction to the imbalance (to much supply and/or to little demand) over-corrects and a new imbalance eventually occurs on the opposite side or in reverse.

Under normal conditions, a market tends to chop sideways. By and large, we do not want to establish a long or short position trade under perceived normal conditions. This is where the IMPA comes into play. It helps us understand when we should, and when we should not be looking at a particular market. We know that under normal conditions, large moves are unlikely (not impossible, just not likely). However, when a disruption between supply and demand begins to unfold, prices adjust accordingly. Using my proprietary statistical indicators, a.k.a., an Upperman Analysis, for measuring commercial activity on the producer and consumer level, we are able to spot certain trading activity which tends to indicate a perceived future disruption in supply and/or demand. How? Remember that futures markets are always adjusting for perceived price changes in the future. The commercial consumers and producers are at the forefront of supply and demand. They will begin certain hedging programs when they perceive future changes in supply and/or demand. Our statistical indicators pick up on any unusual trading activity by the commercial hedgers. Again, always pinpointing the activity of the market participants.

A severe disturbance in supply and demand causes a severe disturbance in price. If supply is perceived as being lower than anticipated, prices will go higher. If demand is suddenly perceived to be greater than anticipated, prices may go higher as well. We measure the supply in the commercial producer category and the demand in the commercial consumer category. The trading activity of these two participants is a reflection of supply and demand. The IMPA identifies unusual trading activity prior to the actual adjustment in price. This provides us with the opportunity to get long or short in front of a significant price adjustment.

My Upperman Analysis of statistical indicators derived from the IMPA and COT data clearly show us when the conditions are right for a price trend or reversal to follow. These are the conditions we wait for, and when we identify them, we profit and profit big. Our one page auto-pilot report provides us with weekly insight on which markets we should be focusing on and which markets may be experiencing unusual commercial producer and/or consumer activity.

In my study of the market participants, I've found that the large moves occur when imbalances in supply and/or demand exist. The current activity in the market (buying and selling) sets the price of a particular commodity out in the future. The markets tend to discount current conditions and adjust for future conditions. It can be anticipated that the large users and producers of the underlying commodities are going to begin making adjustments to offset perceived changes in demands and supply ahead of time. This would be prior to shortages or gluts in supply as well as unusual changes in demand. Often what occurs is the large users and or producers are able to understand or simply identify deteriorating conditions in the physicals before any real deterioration in demand or supply disturbances are noted publicly. The lack of public notification is more or less due to a lack of public interest at the right time. Why? The public tends to be followers rather than leaders. By the time the public finally reacts a good bit of the move has already occurred.

The public (usually by way of the media) tends to focus on current events and not on future perceived conditions. The news is always focused on what is occurring right now while the markets and larger commercial participants are focusing on future perceived conditions. While this may work for TV ratings, it isn’t going to help you identify trends in the futures markets. Once in the news, the actual disruption has already occurred and adjustments on paper have likely already taken place.

Indeed, those paper adjustments will need to be closed out to fully offset the initial preparation for whatever the current disruption might be. Hence, “buying the rumor selling the news.” The public is usually there to take the other side of the trade when its time for the initial adjustments on paper to be closed out. This is also why many contrarian trading methods tend to work. Some people actually place trades based on doing the opposite of what is currently being touted publicly.

Fortunately with tools such as the Upperman Analysis, IMPA setups and COT data, we don’t have to follow the herd or simply follow the money. We can pin down the action of market participants; identify trends; put them in historical context and position ourselves to profit from their likely outcome.

Note: when I say “on paper,” I simply mean futures contracts and not the cash physical.