The New Market Wizards: Conversations with America's Top Traders (17 page)

Also, these approaches are appealing because they play into powerful human tendencies that induce one to trade countertrend or to abbreviate trend-following trades. It’s always tempting to liquidate a good trade on flimsy evidence.

What about cyclical analysis, which is another technique traders use to try to pick tops and bottoms?

There are very powerful scientific methods of cyclical analysis, particularly Fourier analysis, which was invented in the nineteenth century, essentially to understand heat transfer. Fourier analysis has been tried again and again on market prices, starting in the late nineteenth century with the work of the French mathematician Louis Bachelier. All this scientific research has failed to uncover any systematic cyclic components in price data. This failure argues strongly against the validity of various trading systems based on cycles. And, I want to stress that the techniques for finding cycles are much stronger than the techniques for finding trends. Finding cycles is a classic scientific problem.

What about all the various studies that purport to find cycles in price data?

The markets go up and down. So in some loose sense of the word there are cycles. The problem is that you can fit sine waves pretty closely even to purely random patterns. If you allow cycle periods to shrink and expand, skip beats, and even invert—as many of these cycle theorists (or, perhaps more accurately, cycle cranks) do—then you can fit cycles onto any data series that fluctuates. The bottom line is that rigorous statistical techniques, such as Fourier analysis, demonstrate that these alleged cycles are practically random.

Do you believe that attempts to apply artificial intelligence to trading can succeed?

I think that eventually cybernetic devices will be able to outperform humans at every task, including trading. I can’t believe that just because we’re made of carbon and phosphorus there are things we can do that silicon and copper can’t. And since cybernetic devices lack many of our human limitations, someday they’ll be able to do it better. I have no doubt that eventually the world’s best trader will be an automaton. I’m not saying this will happen soon, but probably within the next few generations.

A good part of the academic community insists that the random nature of price behavior means that it’s impossible to develop trading systems that can beat the market over the long run. What’s your response?

The evidence against the random walk theory of market action is staggering. Hundreds of traders and managers have profited from price-based mechanical systems.

What about the argument that if you have enough people trading, some of them are going to do well, even if just because of chance?

That may be true, but the probability of experiencing the kind of success that we have had and continue to have by chance alone has to be near zero. The systems worked for us year after year. We taught some of these systems to others, and it worked for them. They then managed other people’s money, and it worked again. There’s always the possibility that it all could have happened by luck, but the probability would be infinitesimally small.

There has actually been a dramatic shift in the academic view on this subject. When I first started in this business, mechanical trading was considered crackpot stuff. Since then, there has been a steadily increasing number of papers providing evidence that the random walk theory is false. System trading has gone from a fringe idea to being a new kind of orthodoxy. I don’t think this could have happened if there weren’t something to it. However, I have to admit that I find it unsettling that what began as a renegade idea has become an element of the conventional wisdom.

Of course, you can’t actually prove that price behavior is random.

That’s right. You’re up against the problem of trying to prove a negative proposition. Although the contention that the markets are random is an affirmative proposition, in fact you’re trying to prove a negative. You’re trying to prove that there’s no systematic component in the price. Any negative proposition is very difficult to confirm because you’re trying to prove that something doesn’t exist. For example, consider the negative proposition that there are no chocolate cakes orbiting Jupiter. That may be true, but it’s very hard to prove.

The random walk theory has the disadvantage of being a negative proposition. Nevertheless, in the absence of any evidence to the contrary, it might be a plausible theory to maintain. At this point, however, I think there is enough contrary evidence so that any academic who still espouses the idea that the markets are random is not looking at the realities.

In recent years, there has been a tremendous increase in the amount of money being managed by professional traders using computerized, trend-following strategies. Will this proliferation eventually kill the proverbial goose that lays the golden egg?

The question of whether the preponderance of system traders, especially the group of large managers, is spoiling systems trading is difficult to answer because there are two very different kinds of evidence that yield opposite conclusions. First there is the quantitative statistical evidence that systems continue to work. Then there is the qualitative argument that a preponderance of system traders has to change the market in such a way that profit can no longer be extracted in this manner. In other words, the random walk theorists may still have the last laugh. It’s difficult to treat such heterogeneous evidence in a common frame-work so that one kind of evidence can be weighed against the other.

Well, both arguments can’t be right. Which do you believe?

System traders still have an important old ally: human nature. Human nature has not changed. Fortunately, there are still a lot of people trading on their instincts. But there’s no question that the game has become much more difficult.

In evolutionary biology, one of the proposed solutions to the question of why sexual (as opposed to asexual) reproduction is so abundant is the Red Queen Hypothesis, based on the character in
Alice in Wonderland
in whose country you had to run as fast as you could just to stay in place. The idea is that competition is so severe that a species has to evolve as fast as it can just to stay where it is; sexual reproduction provides a kind of evolutionary overdrive. Similarly, there is such strong competition in the systems trading niche that the trader has to develop systems as fast as he or she can to merely stay in place.

Is the implication that the increasing proportion of professionals in the total trading universe will change the nature of the markets in such a way that previously valid systems may no longer work?

I think that’s true. That’s why I’m willing to accept systems with somewhat lower theoretical performance if I think they have the property of being different from what I believe most other system traders are using.

When I raise the point with would-be system designers that much historical research may be invalidated by the changing nature of futures markets, they invariably reply that the solution is to develop systems based on recent data—as if it were that easy. There’s a serious problem with this approach. Recent data has to be less statistically significant than long-term historical data simply because there is a lot less of it. Systems developed solely on recent data are flimsily supported. There’s no way around this basic fact.

If you were starting out again, what would you do differently?

I would concentrate more on money management. To my regret, it was something that I ignored in my early years. Ironically, even though money management is more important than the price model, mathematically, it’s the more tractable problem.

Is there anything unique about your approach to money management?

One drawback to many money management schemes is that they are wedded to the assumption of a logarithmic utility function. Essentially, this model assumes that the increase in people’s utility for additional wealth remains constant for equal percentage increases in wealth. The problem with this model is that it is unbounded; eventually it will tell you to bet the ranch.

There is a technical objection to unbounded utility functions, which is known as the St. Petersburg Paradox. I can give the thrust of it with a simplified example. Suppose you have a billion dollars. If your utility function is unbounded, there has to be an amount of money that would have such large utility that you’d be willing to flip a coin for it against your entire billion-dollar net worth. There’s no amount of money—although there may be nonmonetary considerations (perhaps an extra hundred years of life)—for which a sane person would gamble away a billion-dollar net worth on the flip of a coin. Therefore, there must be something wrong with unbounded utility functions.

We use only bounded utility functions in our work on risk management. The particular utility functions we use also have the desirable technical characteristic of optimal investment fractions being independent of absolute wealth level.

How much do you risk on a single trade? Do you have a formula you go by?

You shouldn’t plan to risk more than 2 percent on a trade. Although, of course, you could still lose more if the market gaps beyond your intended exit point.

On the subject of bet size, if you plot performance against position size, you get a graph that resembles one of those rightward-facing, high-foreheaded cartoon whales. The left side of the graph, corresponding to relatively small position size, is nearly linear; in this range an increase in trading size yields a proportionate increase in performance. But as you increase size beyond this range, the upward slope flattens out; this is because increasingly large drawdowns, which force you to trade smaller, inhibit your ability to come back after strings of losses. The theoretical optimum is reached right about where the whale’s blow-hole would be. To the right of this optimum, the graph plummets; an average position size only modestly larger than the theoretical optimum gives a negative performance.

Trading size is one aspect you don’t want to optimize. The optimum comes just before the precipice. Instead, your trading size should lie at the high end of the range in which the graph is still nearly straight.

How important is intelligence in trading?

I haven’t seen much correlation between good trading and intelligence. Some outstanding traders are quite intelligent, but a few aren’t. Many outstandingly intelligent people are horrible traders. Average intelligence is enough. Beyond that, emotional makeup is more important

I assume you were probably involved in developing the systems that were taught to the Turtles. [See next chapter for background details.]

Yes, I was.

As I understand it, the catalyst for the Turtle training program was a disagreement between you and Richard Dennis as to whether successful trading could be taught.

Yes. I took the point of view that it simply couldn’t be taught. I argued that just because we could do it didn’t necessarily mean that we could teach it. I assumed that a trader added something that couldn’t be encapsulated in a mechanical program. I was proven wrong. The Turtle program proved to be an outstanding success. By and large, they learned to trade exceedingly well. The answer to the question of whether trading can be taught has to be an unqualified yes.

Do you believe that the systems that Dennis and you presented to the Turtles have degraded because there are now twenty new disciples using the same approaches?

With hundreds of millions under management, if they were still trading the same way it’s hard to see how that could fail to be true. However, it’s difficult to say to what extent the Turtles are still trading the same system. I would assume many of them are doing things differently now.

If trading can be taught, can it be taught to anyone with reasonable intelligence?

Anyone with average intelligence can learn to trade. This is not rocket science. However, it’s much easier to learn what you should do in trading than to do it. Good systems tend to violate normal human tendencies. Of the people who can learn the basics, only a small percentage will be successful traders.

If a betting game among a certain number of participants is played long enough, eventually one player will have all the money. If there is any skill involved, it will accelerate the process of concentrating all the stakes in a few hands. Something like this happens in the market. There is a persistent overall tendency for equity to flow from the many to the few. In the long run, the majority loses. The implication for the trader is that to win you have to act like the minority. If you bring normal human habits and tendencies to trading, you’ll gravitate toward the majority and inevitably lose.