how it all started

Back in the late 1970s I enjoyed working with technical analysis on commodity charts. I was a technical junkie; I enjoyed constructing charts and graphs and would spend time drawing and designing my own from information in the local newspaper, including the daily high, low, close and the open, along with open interest for each commodity. Every day I clipped the daily prices from the paper and applied the gathered facts to graph paper to form my charts. On these charts I would then conduct my analysis, which helped me learn the rhythm of the markets. Today this is a lost art among traders. Life and technology has moved on. But, to me, this was fascinating, and hours would slip by without my noticing. In following years I would add the use of Quotrek, a product created by eSignal, to get intraday quotes for charts by means of an FM radio station. Depending upon where you were you had to adjust the antenna to obtain decent reception! Five-minute charts were easy to construct, as all you needed was Quotrek, a pencil and a piece of paper to chart, and a watch. The watch was used to time the bars as each number represented five minutes, so when it went from one to two that was the end of a five-minute period. Trend lines were drawn and the analysis, like momentum or stochastics, was done in longhand and added to the chart. It sounds crude today, but it sure gave you a feel for the rhythm of the markets. My first attempt at a live trade was a system I tested for some time, which

fig. 1.1

this chart in cotton is similar to my first trade. i did not use the Current relative Strength (CrS) momentum indicator, but i did study it and observed how it provided information.

Source: Commodity research Bureau

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

appeared to work very well. On paper it did just fineso I was eager to try it live as I believed then, as I do now, that systems must be tested live, in the heat of battle, and not after the fact. The system was simple and straightforward: Take the closing price on a daily chart, long and short, when it closed above/below a 10-day simple moving average. The filter I used was a 40-day simple moving average, and it had to be sloping in the direction with the 10-day average. Stop placement was close above/below the 40-day average, therefore creating a trade with trend. This system then had all the ingredients of the perfect trade: entry, trend following, money management and exit. Even today, I would not change a thing on this plan. My first trade was in corn (figure 1.1 on the previous page looks similar to my corn trade). Corn moves slowly, is great to trade and works well with the system. I called my broker and put in an order to buy and at the same time put in an order for a stop. Trading my plan, I submitted my order as, Buy five Christmas corn at $2.20 on a buy stop, and an order to sell on sell stop five Christmas corn at $2.15 when filled on buy stop. This was easygetting the order correctly to the broker was the hardest part. So, now I had an order in to buy. This was exciting! Adrenaline flowed! I called the broker five minutes later to ask if it was filled. Nothing yet, was the answer. OK, no problem. I asked him what the last corn price was. It was trading a couple of cents below my buy stop, so we were golden! About a half-hour later I called the broker back and I was told Got filled on stop to sell but no fill on buy but it traded through buy price so should be filled on buy. What?! Filled on sell stop first and dont know fill on entry. My heart beat faster and faster. This could not be happening. It never happened on the paper testing. This was an eye opener for me as I received, a couple of minutes later, my buy stop order fill that was triggered before sell stop trigger. Net was, I went long corn before it reversed and I was stopped out for a loss. Deflated, confused and sadall these emotions came out. On my first live trade I learned so much. Over the next few years trading I recalled, many times, the harsh lessons learned about live trading from that heart-stopping experience. Today traders complain if the fill is not made within a second or two! This event set me on my way to find a better systemsomething that

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

worked! Never mind that it was only one tradeI had to tinker and make it better right away. For many years I used every setting on every indicator that was available, as well as combinations of those indicators, in the search for the one unique thing that would help me conquer the markets. However, I always found myself returning to momentum indicators. Looking again at the posted chart and the momentum indicator, current relative strength, I noticed a turn in it before prices; it also formed divergences very well. From that point on most of my technical analysis was on momentum and momentum indicators. Momentum gave me a definite edge and I pursued that aspect of analysis. Notice the turns and the divergences on figure 1.2 below; this was key for me.

What makes trading the CCi revolutionary?

As a technical analysis junkie, I started to see patterns develop on momentum type indicators. In the late 1970s there were very few books or magazines on trading. And of course there was no Internet available for research. Around this time I came across an article in Commodities magazine, currently known as Futures, on the Commodity Channel Index, written by a person named Don Lambert. After reading this article and starting to use the CCI I noticed shapes, or

fig. 1.

1981 no.  Cotton With divergenCe rising 10 and 0 Moving averages (Mas) and rising positive rising 0 Ma and positive CrS rising 0 Ma and positive CrS declining 10 and 0 Ma and declining negative CrS CrS crosses zero base line Closing price crosses 0 Ma after a prolonged move 10 Ma crosses 0 Ma CrS indicates if trend is gaining or losing strength CrS indentifies strongest and weakest acting commodities Current charts of 10- and 0-day (composite) weighted moving averages and momentum oscillators for each of 6 commodities.

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

patterns, showing up on the CCI every day, regardless of what time frame. This is where the revolutionary part comes into play. I found that there was nothing written or documented about the patterns that formed on an indicator. Again, all I could find was information about price bar patterns. The longer I tested and applied the CCI, the more it showed that the patterns worked with trading, not 100 percent of the time, but with a good probability of success. I was then faced with creating names for these patterns. There was nothing written about patterns on an indicator so I went ahead and named them myself. I had wanted to find an approach that had never been used or traded before and this was it! I chose Ghost as the name for my first identified pattern because it looked like a ghost, with the head and shoulders raised like a Halloween phantom. Price bars have the Head and Shoulders and now we have the Ghost on an indicator. It was completely revolutionary compared to what had come before. The next pattern I tried and tested became known as the Zero Line Reject (ZLR). The more time I spent evaluating and trading the ZLR and watching ONLY the CCI the more I realized that the patterns had a psychological aspect. The ZLR showed the buy the dip and sell the pullback mentality. Everyone tells you to do thisbut no one tells you when to do this. The CCI ZLR did! From that point on I found that every pattern I developed told a story. Perhaps you can now see why trading the CCI along with my patterns is groundbreaking. Many people are using it and it has helped a considerable number of traders around the world. While working with the CCI, I found myself watching only the CCI and not the price bars. This led me to stop monitoring the price bars altogetheralong with those mind games they play on all traders. I teach this now to traders, both experienced and those new to trading, and tell them, We dont need no stinking price bars! This method was completely new and totally different from standard techniques, and those traders using it discover a big difference in their trading. Trading becomes less stressful and using the CCI allows them to make good trading decisions. Another factor that is revolutionary to technical analysis with the CCI is the use of the key value lines on the CCI panel. The 100 lines and the zero line will show support or resistance.

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

Support and resistance signify important points in time where the forces of supply and demand meet. In the trading markets prices are driven down by excessive supply and up by intense demand. Supply equals selling (the bears) and demand equals buying (the bulls). The zero line is major support or resistance at this time; I mean real time, not projected time. Many things happen around the zero line and if you watch the CCI you will see this. The CCI tells a story and gives great insight as to what is happening. The 100 lines are minor support and resistance and again, by observing the CCI, you will notice this. No other indicator can you show this. Add the Turbo CCI to the trading panel to complete the story of what the CCI can tell you but also note the warnings it will give you. No other indicator has this capacity or can show you what is going on in the trading or investment world. This is the reason that the CCI has proven exceptionally helpful for many traders.

a new, revolutionary way of trading the CCi

Using the CCI is a revolutionary way of trading my patterns on an indicator. The CCI can be used in any market and in any time frame, including forex, the stock market, indexes, futures and commodities. Many charting packages now have Woodies CCI patterns incorporated into their charts, and it is being added to others all the time. These include eSignal, Trade Navigator by Genesis, Sierra and Ninja. There is much written about the CCI and how I teach it, as more and more traders use it. For example, in July 2006, the basic CCI document was downloaded from the Woodies CCI Club website 38,000 times. This phenomenal growth in interest in the CCI continues unabated, and many traders are taking advantage of our marvelous website. There is so much to be learned from this ever-expanding goldmine. The information available includes access to the Trading University, which holds more than 10,000 articles on trading contributed by some of the 3.5 million people who have visited the site to gain knowledge and to offer nuggets of wisdom to help their fellow traders, new and experienced there it is again: Traders Helping Tradersour constant mantra.

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

the CCi comes to life

Donald Lambert developed a fascinating indicator called the Commodity Channel Index in 1979. Don was not a trader but a mathematician and statistician who was brilliant at developing programs for trading. His business, Lambert Programming Service, supplied programs to traders using two calculators, the HP-41 and the TI-59. He was the authorized programmer and distributor for a number of well-known trading system originators. The CCI was developed to test the computing capabilities of a portable calculator, not to trade with

the cci can be used in any market and in any time frame.

it. In fact, Don has never used the CCI and does not intend to use it. Interestingly, however, he wrote the article for Commodities Magazine that I had seen on how he thought it could be applied and traded in the markets. Don and I have become well acquainted over the years as I have developed patterns on the CCI and grown my own Traders Helping Traders business. He has given me a lot of his original work because of the techniques I have created with his indicator. My strategies with the CCI are completely different from what Don originally thought was possible, and he is pleased the CCI is helping traders. Don has delivered excellent presentations at my CCI seminarshe is a great speakerand I am sure Don will be seen at future CCI events. I have included much of his work in the following pages, including the article he wrote back in 1980 that piqued my interested in the CCI. That article is still listed today in most charting packages, along with the calculation of the CCI as Don originally did it. Calculations are also listed for those who prefer to use spreadsheets. There is not much written about the CCI, other than Dons work, much of which has been reprinted here. I will explain and identify the setups that are traded now, which have evolved over the years into something completely different from the strategies and trades of the early years.

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

CoMModity Channel indeX: tool for trading CyCliC trendS

by donald r. lambert, october 1980
(this article first appeared in Commodities Magazine, october 1980. reprinted with permission of futures Magazine,  S. riverside plaza, Suite 60, Chicago, il 60606.)

Many commodities exhibit some type of cyclical or seasonal price pattern. But, the commodity trader still faces the problem of detecting when these regular price movements begin and end because climate and other real world conditions may affect their timing. One method that can help spot these turns is the Commodity Channel Index (CCI), a recently developed index which is somewhat akin to a standard score in statistics. The CCI doesnt calculate cycle lengthsyou must determine them yourself or rely on an advisory servicebut is a timing tool that works best with seasonal or cyclical contracts. To be useful in cyclical markets, an index must examine current prices in the light of past prices but must not allow data from the distant past to confuse present patterns. For this reason, the CCI uses a moving average rather than an exponentially smoothed average as a benchmark against which to measure current prices. The comparison of current prices to moving averages

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

solves one problem by providing a moving reference point. But, it leaves another problem for the trader: while some commodities typically move only a few cents each day, daily moves in others might be hundreds of cents. Rather than develop separate rules to determine each commoditys fluctuations, some standardization technique had to be found. A very simple solution to this problem would be to divide the difference between the current price and the moving average by the daily limit for the contract, thus producing a measure whose dimension would be contract-independent. But, a final considerationthat a given price movement of a contract will not always have the same significancerules out the daily limit move as a divisor. A divisor should adjust to price action. It should be relatively small when prices are oscillating but should become larger when a breakout occurs. In short, the divisor should reflect not only a contracts possible trading range but also the contracts actual

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

recent trading pattern. In statistics, such numbers affected by both the size of data and fluctuations in data are called measures of variability.

not by hand
Most current trading methods were worked out by hand originally. But with the availability of low-cost, high-capacity programmable electronic calculators today, systems originators no longer are bound by pencil-pushing limitations. Using a TI-59 programmable calculator to test various divisors, I chose the mean deviation as the divisor for the Commodity Channel Index. It has never been calculated by hand, and for all practical purposes, cannot be done by hand on a dayto-day basis because of the time-consuming nature of the methods used. The first two steps are easyfinding each days typical price by averaging the high, low and close and computing a moving average of those typical prices. Step 3

10

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

is the most difficultcomputing the mean deviation for the number of days desired from the new moving average each day. Once the mean deviation is known, however, you can compute the CCI quickly. The use of a 1.5 constant in the CCI formula scales the resulting CCI value so 70% to 80% of the random fluctuations fall within a +100% to -100% channel. If the CCI goes above the +100 line, that is a signal to establish a long position. When the CCI drops below the +100 line, the long position is closed out. The same techniques apply to short positions at the -100 line. You can modify the CCI to set your own parameters. Probably the most critical factor in using the CCI is the choice of data base length. Too short a data base will produce whipsaws as the index interprets daily price fluctuations as being cycle tops or bottoms. Too long a data base, on the other hand, slows response time to such an extent that breakout indications are given later than is desirable. Comparison studies were done on the interaction of data base length and cycle length for a theoretical perfectly cyclic contract (PCC). While the 10-day data base CCI detected cycle tops for various cycle length PCCs well, its ability to detect a breakout suffered for short-cycle length PCCs [see first table on previous page]. Similar studies using 5-, 15-, and 20-day data base CCIs [see CCI profit efficiency table on previous page] suggest the data base should be less than one third of the cycle length to produce a reasonable level of theoretical efficiency. A rather remarkable result was finding that the CCI always gave for all PCCs an exit signal either at or before the extreme price, never after the extreme price. While the 5-day data base CCI has the highest theoretical efficiency level for all cycle lengths studied (table 2), it probably would be susceptible to whipsawing. Based on that assumption, 20 days was set as the standard length of the data base in the TI59 calculator program although any period between 5 and 25 days can be used.

Donald R. Lambert has degrees in mathematics, statistics and accounting and has been teaching these subjects on a private basis for 25 years. Lambert is not a commodity trader himself but runs a programming service from his home in Los Angeles. He is the authorized programmer and distributor for a number of wellknown trading system originators.

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

11

the following is a letter i received from don lambert when i started corresponding with him after reading his article in Commodities Magazine:

Dear Trader: The Commodity Channel Index is an original development of the Lambert Programming Service. It has never been traded. Tests on historical data have produced mixed results, since in each case a trader would have to decide whether to use it for trend following or channel trading. Attached in a graph (September 1979 oats) in which it could have worked for either. I have assumed that the trend follower takes a position when the index passes out of the channel and closes out when it re-enters the channel. The channel trader, on the other hand, was assumed to take a position when the index enters the channel and to close out when it exits the channel. For purposes of this example, 100 and -100 were taken as the channel boundaries. Results, with actions taken on the openings, might have been as follows: Trend Follower: DATE Jun 8 Jun 27 Jul 13 Jul 19 Jul 20 Aug 7 ACTION BUY SELL SHORT BUY SHORT BUY PRICE 162.25 178.50 166.00 167.75 163.00 139.75 RESULT +16.25 -1.75 +23.25 +37.37

TOTAL (ignoring commissions)

1

c W ho ao pdti ee r 1B :o t h e Crhea p t e r it i n a e y Way o f t r a d i n g t h e C C i s ok: volut o tl r

 Channel Trader:

(end of data base)

DATE Jun 27 Jul 13 Jul 19 Jul 20 Aug 07 Aug 22

ACTION SHORT BUY BUY SELL BUY SELL

PRICE 178.50 166.00 167.75 163.00 139.75 150.00

RESULT +22.50 -4.75 +10.25 28.00

TOTAL (ignoring commissions)

As with any system tested only on historical data, actual results could have been far different due to market conditions, and these results do not necessarily indicate future expectations. Yours for good trading, Donald R. Lambert 1979 oats, September contract

c h a p t e r 1 : t h e r e v o l u t i o nW oroyd Way B o f kt: rCahdai n t e t ht i t l ei a ies o pg r e CC

1