Stock Market Wizards

by Jack D. Schwager

  I cold-called Eastman Kodak. That initial call ultimately led to their opening a $50 million account—the largest investment ever in managed futures. They eventually upped their investment to $250 million.

  What did you know about managed futures?

  Nothing, but I did know enough to realize that it was a waste of time to call individuals and that it made a lot more sense to call institutions.

  Then how did you sell Kodak on the product?

  I told them, “Here is an investment that has no correlation with the stock market and has been compounding at about 30 percent per year.” The Kodak account started me toward financial independence.

  After the Kodak sale you must have thought: “This is really easy!”

  I expected the money to pour in.

  Were you successful at opening other accounts?

  We tried to open other institutional accounts, but nothing happened. We basically had one account. No other institutions stepped up the plate.

  So, on your first sales call, you landed a $50 million account, and then you never made another sale again.

  It’s hard to believe, but it’s the honest-to-God truth. The Kodak account was my only source of income.

  Still, given the size of the account, you had to be doing pretty well.

  We were making a lot of money off the account, but the problem was that it was a typical managed futures account—up-and-down, up-and-down—it was sickening to watch. The traders would make money, and then they would give it all back. I was concerned about losing the account because of all the volatility. So I started looking around for something else to do.

  Sometime around 1993, I became interested in a stock market newsletter written by a guy in Texas. He put out recommendations on mutual sector funds and had a good track record. I called him up and suggested that we do a fund. He agreed, and the fund was launched in September 1993. He was the trader, and I raised the money.

  Had he traded before this fund was formed?

  No, he was just a newsletter writer. This was his first experience with trading real money.

  Hadn’t the idea of trading occurred to him before?

  I think he was somewhat conservative. He had a good position at IBM, which he was reluctant to give up. He had been writing the newsletter on the side. I convinced him to leave IBM. In the first ten months I raised approximately $10 million. After the first year, he was up about 9 percent with a lot of volatility. I realized that this was not for me—the equity swings were just too volatile relative to the mediocre returns being realized.

  By late 1994, I had become completely disenchanted. At the same time, I had begun doing my own research on mutual fund timing and thought I could do better. The trading manager and I agreed to split up. He kept the individual managed accounts, and I took over management of the partnership account.

  You said you began doing research. Had you developed a trading method by the time you took over as the fund manager?

  No, I didn’t have enough confidence in my research. I knew I wasn’t quite there yet.

  Then what was your plan for trading the fund?

  I didn’t have any great plan. I just knew that what we were doing wasn’t working. I had enough confidence in my abilities to believe that I could come up with something better.

  So your trading method was still a work in progress at the time you took over the trading responsibilities.

  Yes.

  Did you consider delaying the split with your partner until you had developed your own trading strategy?

  No, I knew I would come up with something. There was absolutely no doubt in my mind. I had never failed to succeed at anything that I put my mind to, and this was no different.

  Still, you had never traded successfully.

  The characteristics of being a good trader or investor are very similar to the traits needed for success in general. I think it would be very difficult to find someone who was not successful at what he was currently doing, put him in a trading position, and make him successful. I don’t think that is going to happen. The same qualities that make you a successful person in whatever you’re doing are going to make you successful in trading. You have to be very decisive, extremely disciplined, relatively smart, and above all, totally independent. I have those traits. Therefore, when I decided to become a money manager, it didn’t require a leap of faith to believe that I would be successful.

  Since you hadn’t fully developed an alternative trading method when you began trading the fund, how did you make your trading decisions?

  It was a joke. I didn’t know what I was doing. I did what everyone else does. I looked at a chart, and if it looked strong, I bought it.

  How long did this go on?

  For most of the first quarter of 1995. I was lucky to finish the quarter up a few percent. By March 1995, I had systematized my approach and felt confident that I had come up with something that would work. I implemented an embryonic version of what I do now.

  That implies that you have changed your system very substantially since you first started. Were these changes a consequence of ongoing research, or were they triggered by your trading experiences?

  I had several important events in my track record that caused me to change significantly from what I had started out doing. I had a very good first year. I finished 1995 with a 58 percent gain and no losing months from the time I adopted my systematic approach.

  In January 1996, however, I found myself down about 5 percent by midmonth. That may not sound like much to most people, but to me it was a huge amount. Because of that drawdown, I spent an enormous amount of time doing research on the computer and ended up making very significant changes to my methodology.

  Everything went along well until late 1996, when my trading results went relatively flat. For the fourth quarter of 1996 and the first quarter of 1997 combined, I was up only a little over 1 percent. This was definitely not what I was looking for. I realized that I had to make some changes. During that period, I was on the computer all day, almost every day. In March 1997, I implemented some very significant changes to my systems. Since then, the performance has been quite good.

  Although my systems have been unchanged since then, over time I realized that I could combine my systems with my experience. Now my systems tell me what to do, but there is also judgment involved. This judgment doesn’t necessarily make me more money, but it does reduce my equity swings. I usually err on the side of caution if I lack conviction on a trade.

  Can you give me an example of how you use judgment?

  There is never any judgment whether to buy or sell; the only judgment is how much to buy or sell. The problem with system trading is that it doesn’t tell you how to trade your portfolio; it just gives you buy and sell signals. I trade several different systems, each based primarily on one indicator. I might have a system that has been performing extremely well give me a buy signal, but I may decide to take a smaller-than-normal position because other systems are giving me contradictory indications.

  What is another example of judgment causing you to deviate from the strict signals of your system?

  Let’s say the market has been trending up for a while, my systems are long, and I’m making a lot of money. Although everything may look great, I get uncomfortable when my equity line starts going above its long-term uptrend. I am likely to cut back my position size, anticipating that the equity line will come back to the long-term trend. Judgment like that saves me money rather than makes me money.

  Judgment is also important in deciding which systems I use. Interestingly, the systems I used a few years ago are not doing particularly well anymore. Somehow I’ve been successful in changing so that I’m usually trading the best systems. I can’t tell you how I’ve managed to do it. I guess it must be intuition.

  If you stop trading a system because it shows some deterioration, do you sometimes go back to using it several years later?

  No, because I replace inferior systems with superior
systems. There is a reason why I replace trading systems, and the reason is that I have a better idea. I still keep an eye on old trading systems, but I won’t use them.

  Doesn’t it sometimes happen that a discarded system begins performing better than a system you are currently using?

  It probably will happen at some point, but it hasn’t happened yet.

  Do you trade individual stocks?

  No, although it is very likely that my systems would also work on stocks. In fact, that’s my next research project.

  So, what is your trading vehicle?

  Mutual funds, but I’m not a market timer. Let’s discriminate between a market timer and what I consider myself—a market reactor. A market timer says, “The market is too risky here. I think the Dow is going down to 8,000 during the next three months.” They have a view about what is going to happen. They prognosticate the market. I do not attempt to prognosticate the market. I react to what happens in the market.

  Your actions, however, will be the same as a market timer. You will switch back and forth between a mutual fund and cash, based on the timing signals of your systems. Isn’t that the same thing as a market timer?

  The actions may be the same as a market timer, but the thinking is completely different. I make no predictions. I have absolutely no idea what is going to happen [he laughs].

  Why are you laughing?

  I’m laughing about the people who do make predictions about the stock market. They don’t know. Nobody knows. I don’t think anybody has any idea what is going to happen in the stock market.

  Does your own performance depend on the mutual funds you choose to trade?

  Only to a very limited extent.

  Do you trade mutual funds that represent the broader market?

  I have tested my systems on marketwide funds, and they work well. But I usually prefer to go after a smaller area of the market. I’m looking for funds that would have a bit more zing on the upside, and the S&P is not zing. Therefore, I’m much more likely to trade something like a technology fund than a broadly diversified fund.

  I don’t expect you to reveal the systems you are currently using. However, are there systems that you developed in the past and that worked for a while, but are worthless to you now? At least that would provide an illustration of what a system idea that worked for a while looks like.

  I can give you an example of something that might not be too distant from what I used to do by describing my perception of Gil Blake’s system. [Blake was a trader interviewed in The New Market Wizards.] Gil’s approach was to follow different sectors, and if on a given day, a sector had both above-average volatility and above-average return, it would be considered a buy signal for that sector fund, or a “green light,” to use his terminology. Then he would hold the long position in that fund until his sell condition was met, which might have been a down day, or the passage of a specific number of days following the buy signal, or some other liquidation condition.

  That system provides a good example of the kind of thinking that I do. There is no reason why you couldn’t implement that type of system today. Although it wouldn’t do remotely as well as what I am using now, it would probably still work to some degree.

  Did you have that type of idea before you read the Gil Blake chapter?

  No. Reading the Gil Blake chapter was a key turning point for me. Although what I do now has nothing to do with what Gil was doing then, it at least helped me get to the point where I could start doing research on the computer.

  Did you ever talk to Gil Blake?

  Yes, I called him when I first started managing money and said, “My name is Steve Lescarbeau, and I just wanted to tell you that you’re the reason I’m in this business.” He groaned, “Oh God.”

  Yes, I could imagine how many times he had heard that line. If you had not read his chapter, would you have ended up in this business?

  I don’t know; it was that important.

  Does your original trading system—the one inspired by Gil Blake’s interview—still work?

  It works, but it has degraded a lot.

  Do you think this might be a temporary phase and that in the future it might start working very well again?

  I doubt it.

  Can you see yourself ever using it again?

  No.

  If you don’t expect to ever use it again, and it’s not related to what you are doing now, is there any reason why you couldn’t talk about it more specifically?

  Well, you never know [he lets out a long laugh].

  Are you still in the process of trying to improve what you are doing?

  Absolutely. I’m trying to, but I don’t know if I’ll be able to. It’s hard to improve on 60 percent a year, * but I’ll be happy to maintain it. I’m constantly concerned that it is going to go away. In fact, I know it will. If you come back a year from now, I’ll probably be doing something different. I’m sure what I’m doing now won’t work as well as it has up until this point.

  You are implying that systems have a life span.

  There is absolutely no doubt about it. There is no way anyone could convince me otherwise. Systems definitely have a life span.

  Why do you think that is?

  I think it’s because eventually enough people figure it out. When too many people jump on the bandwagon, the market takes it away. That’s why I would be very skeptical about anyone being able to buy a trading system that worked—that is, a system that made money with an acceptable level of risk.

  If you develop a system that you have thoroughly tested and truly believe works, don’t tell anyone about it. Use it, because it’s going to go away at some point in time. Understand that it won’t last forever, and work on coming up with something different for when that happens.

  I’m always concerned about people figuring out what I do, because I know if that happens, it’s going to stop working. For example, the “January effect” is gone. [The January effect is the tendency for small capitalization stocks to outperform large capitalization stocks during January—a pattern that until 1993 had repeated in over 90 percent of all years since the mid-1920s. Then the pattern failed six years in a row. Lescarbeau is implying, quite plausibly, that the January effect’s increasing publicity triggered its own demise.]

  If too many people are using the same system, what mechanism in the marketplace causes the system to self-destruct?

  I can’t answer that question. It could just be a matter of too many people on the same side of the trade at the same time. Everything I have experienced tells me that systems have a life span, and not a terribly long life span.

  That speaks to the death of systems; what about the birth of systems? Will systems start working at some point in time, say 1994, and then stop working a few years later? Or if you tested the systems you are currently using over the past twenty or thirty years, would you find that they worked over the entire time span, but it’s just a matter of your not finding them until more recently?

  Usually when I discover a system, it’s been working all along. Having said that, though, I find that the systems that have done the best in the most recent past also tend to do the best in the immediate future. Therefore I tend to lean on the systems that have done the best very recently.

  You say that systems have a limited life span, but by your own admission, the systems you are using have worked for over twenty years. Why couldn’t they work for another twenty years?

  I understand where you are going with that question, but I don’t agree with the conclusion. I don’t buy it because there is just a lot more money pouring into the markets. The best example is the commodity markets. When we sold the managed futures account to Kodak, the traders managing the account had systems with great track records; these systems had been averaging 40 percent for fifteen years. They said there was no way that these systems would stop working. Well, they did. They stopped working because too many people started using similar systems.

  Another classic example i
s O’Shaughnessy. His book What Works on Wall Street was terrific; it was well written and well researched. The performance of his funds, however, has been less than stellar.

  What do you consider “less than stellar”?

  [At this point, Lescarbeau looks up the performance of O’Shaughnessy’s funds on his computer screen. He checks two of the funds and finds that they are up 43 percent and 46 percent. Although this doesn’t sound too disastrous, during the same time period (late 1996 to mid-1999), the S&P 500 was up 89 percent. So, these funds made only about half as much as the S&P 500, which is representative of the benchmark they were designed to beat.] Great book. He tested his strategies all the way back to the early 1950s, but they don’t work.

  So, even though his strategies worked for over forty years at the time the book was published, they have stopped working in recent years.

  You know what? Had he not published his book, they might well have continued to work. He should have just managed money and not published his book; of course, if he hadn’t published the book, he probably wouldn’t have raised any money.

  Your premise is that his strategies stopped working because too many people were following the same ideas.

  Exactly. The most important message I can give anyone who reads your book is that if you have a great idea, don’t talk about it.

  Some people I have interviewed say, “I could publish my system in the Wall Street Journal and it wouldn’t make a difference.” I take it that you don’t agree.

  I’ve read statements like that, and I couldn’t disagree more.

  You feel that if you described your system in The Wall Street Journal, it would stop working.

  It would be over. Tomorrow [he laughs]!

  At one point, you had investors, but you no longer do. What happened?