AR The great thing about science is, your bad ideas don’t count against you. Trading is a little bit different. Your bad ideas can lose you money.”
Have you always been interested in trading?Not really. When I got married for the first time, I was at graduate school at Berkeley, and we had some money from wedding gifts, and I thought I’d invest it. So I picked a couple of stocks, and they lay there like lumps. I complained to my account executive at Merrill Lynch, and he said, “Well, if you want action, you might invest in soybeans.” Of course, he was talking about investing in soybean futures. I bought two contracts and proceeded to make a lot of money over the next couple of days, only to lose it over the couple of days after that. I started going in earlier in the morning, when the markets opened in Chicago, for a couple of weeks and quickly realized that I was going to either write a thesis or trade soybeans, because I couldn’t do both. I elected the thesis and closed out the soybean contracts with, as it turned out, a small profit, but it was obviously 100 percent luck.
What role does luck play in investing?
People underestimate its importance except when things go badly. Then they’re very happy to ascribe it to bad luck, which it may be, of course. One has to recognize that luck plays a meaningful role in everyone’s lives. You get born to decent parents in a good part of the world, and you’re way ahead of the game. And that certainly doesn’t have much to do with skill or hard work. In my case, I was lucky to collaborate with some very good people when I was doing mathematics. I was also lucky in my choice of partners at Renaissance.
How has luck figured into the success of Medallion?
We started off in the ’70s trading fundamentally — we’d read the newspapers, tickers, news wires, and come to conclusions. And we were pretty good at it. The markets were particularly easy. The dollar was going down like a stone, and the long-term interest rates were going up to the moon. So if you would short the dollar and short the bonds, you’d do pretty well; we made a fair amount of money that way. But I began to look at the data and conclude that maybe one could model this. I brought in a friend of mine, Lenny Baum, who was the best code cracker in the world when we worked for the NSA. I asked if he’d like to work with us for a while and try to make some models. I think I was pretty lucky to know that guy, and indeed, we made some models. Now, it turned out that Lenny didn’t really want to use these models because he was doing so well just guessing which direction the markets would go. But the die was cast. By the late ’80s we were committed to models.
We made quite a lot of money trading fundamentally. But I didn’t know how to build a business based on traders. I didn’t know how to judge them. If a guy was extremely good, it would take a while to prove that, and by the time he’d prove it, he would probably be too rich for me to hire. But I did know how to hire scientists and mathematicians. And so we decided to build a business along those lines.
What makes a good scientist?
Creativity is the most important thing. You need to be pretty smart in the conventional way, but science and research are about being creative. It’s about discovering something new, something that wasn’t known before. And you don’t find that out by reading books or looking in the library. You need creativity, and you need ideas.
Lenny, the famous Lenny, would say to me: “Bad ideas is good. Good ideas is better. No ideas is terrible.” And that’s the way, in a sense, that science and research work. You have to generate a lot of bad ideas for every good one. And the great thing about science is, your bad ideas don’t count against you. Trading is a little bit different. Your bad ideas can end up losing you money.
Don’t you test your ideas first?
Everything’s tested in historical markets. The past is a pretty good predictor of the future. It’s not perfect. But human beings drive markets, and human beings don’t change their stripes overnight. So to the extent that one can understand the past, there’s a good likelihood you’ll have some insight into the future. It’s worked for us.
Do you ever override your models?
There are no human overrides built into the models. But every year or two, if there’s some event that looks to us like it indicates more volatility in the markets than the computer could have guessed, we’ll override the system.
During the second week of August last year, many quant shops, including Renaissance, experienced huge, unanticipated losses as a mass wave of selling hit the equity markets. Was that a time to override the models?
We did lighten up for those few days. Volatility was huge and, I might say, in the wrong direction. So volatility, plus the fact that we were losing money at a rate that we had never seen before, inspired us to lighten up, which we did for a couple of days. And then it seemed to be over. We stopped lightening up and started making money again.
What lessons should investors learn from last August?
The positions that collapsed those few days in August all came back, and came back rather quickly, almost certainly being taken on by a different set of players. This told me two things: first, that there’s more overlap among quant investors than I might have guessed. So we may think — and we have some reason to think — that we’re the best at what we do, but that doesn’t mean that everything we do is completely different from what everybody else is doing. Second, that our positions were perfectly sound and that this particular incident was brought about by a rush for liquidity on the part of some participants, not by the frantic desire to dump positions which no one else would want.
Renaissance seems to borrow a lot from academia.
The best science is done in an open exchange of ideas. So when a new researcher comes in here, we want him to learn everything about the system. Learn about every predictor that works, every way that we do things. And learn about things that have failed. Once he’s up to speed, he’ll get his own ideas. What you don’t want inside the walls is secrecy. If people don’t talk to each other, it’s just that much less likely that good ideas will have their day.
Does that mean that everybody within Renaissance knows exactly how you make money?
Well, no. Not because we refuse to tell them, but because they’re either not particularly interested or not able to think about it.
What would you consider to be your greatest legacy?
Renaissance. I was the founder, and it’s achieved a good success. I’m also proud of some of the mathematics that I did. And now that I’m involved with philanthropy, particularly in autism research, I’m proud of the progress that we’re making. I’m also pushing hard for Math for America.
Is the U.S. really that far behind when it comes to math?
We have this curious problem in the United States. Generally speaking, teachers of math and science don’t know math and science, particularly at the high school level. Obviously, some teachers do, but it’s fewer and fewer. To correct the problem, we have to pay teachers more and make their jobs more attractive. If we don’t, the U.S. is not likely to maintain any kind of economic leadership in the next 20 or 30 years. Praying is not going to work.
— Interview by Michael Peltz
No comments:
Post a Comment