The World According to Marty O’Connell
Although statistics are hard to come by, it’s likely that nobody has taught more people about options than Marty O’Connell. Since 1977, thousands of exchange locals, over-the-counter traders and hedgers have passed through the Chicago offices of O’Connell & Piper Associates. O’Connell was a market-maker on the Chicago Board Options Exchange from 1977 until 1984, before branching out into managed futures accounts, management consulting and seminars. He spoke with editor Joe Kolman in August.
Derivatives Strategy: How did you get started in the options business?
Marty O’Connell: I got my start in 1977 as an independent market-maker on the Chicago Board Options Exchange. What attracted me to the options business more than anything was the mathematics of it. I liked the Black-Scholes model and was more comfortable with it than I should have been at the time. On the other hand, after I was in the business for no more than a couple of months, it occurred to me that there was something else going on in addition to the math. Option-trading brings out certain kinds of behavior—mostly human weakness rather than strength. Over the years, that’s been even more fascinating to me than the math.
DS: Why do people seem to make so many mistakes with options?
MO: The options business is mostly about managing three dimensions of exposure: There is usually some exposure to the underlying price, to the passage of time and to changes in implied volatility. In other words, if I have an option position, my P&L will probably have a lot to do with changes in these variables—individually and in combinations.
The rest of the complexity comes from the fact that my P&L will almost always respond in a nonlinear way to each of these variables. On top of that, there often are other minor dimensions of exposure associated with a particular commodity and a particular situation. Most of the trouble people get into with options comes from a lack of appreciation for the multidimensional, nonlinear character of the products—or from an attempt to avoid it.
DS: They want to simplify, and that’s where they get into trouble?
MO: Sometimes they try to construct a situation in which they get the benefits of the multidimensional, nonlinear nature of options without the problems that come with it.
DS: Do you have an example?
MO: Sure. People may simply focus on one variable and imagine that that’s where the action is going to be. A dealer, for instance, might want to bet on a change in implied volatility, because he thinks implied volatility is going higher or lower, or because he thinks there is going to be some shift in the implied volatility curve, and he wants his P&L to reflect the correctness of his view. But in many cases, the position’s P&L will respond to the change in implied volatility in a nonlinear way. Also, he may have to live with the fact that the underlying price movements or time decay might have more of an effect on his P&L than the accuracy of his view of implied volatility. In other words, he might want to bet on a change in implied volatility, but most of his P&L might depend on the actual volatility of the underlying.
Another oversimplification is the over-reliance on Greek letters for the risk management of options. This isn’t to say that the Greeks aren’t useful tools. Of course they are, but there are two problems. One is that they focus only on one variable at a time. Delta and gamma are measures of an exposure to underlying price movements, assuming none of the other variables is going to change. Theta and vega serve similar roles with respect to changes in time and implied volatility. Each of these Greeks offers a one-dimensional view of a three-dimensional puzzle. What’s more, a Greek letter considers only incremental effects—it doesn’t show your exposure to a discrete change in any of the variables. In the real world, instead of a small change in a variable we often wind up with a big change, perhaps in more than one variable. And the Greeks don’t even begin to address those exposures.
I’ve done a lot of trading in my life. I’ve had winners and losers, and every once in a while I’ve gotten beaten up. When I’ve lost a lot of money trading options, it was never because of an incremental change in anything. It always happened because there was a big change in the underlying price, time, implied volatility or some combination of those things.
DS: You are careful to distinguish between the effects of actual volatility and the effects of changes in implied volatility. Are traders often confused by these dynamics?
MO: It is amazing how often traders—even experienced traders—get hurt by sloppy thinking about volatility. For example, in January 1991 we were coming up to the Gulf War. You might recall that U.N. forces weren’t going to start bombing Iraq until at least January 15. As we got closer and closer to that date, financial instruments moved around a little bit but implied volatility kept getting higher and higher, because traders thought that as soon as the bombs dropped—or didn’t drop—there would be a lot of movement in financial commodity prices.
When the bombs finally dropped, we did get significant volatility. Some financial instruments moved a lot, but the options’ implied volatilities collapsed rather quickly. Some people who thought they were betting on high actual volatility were really more exposed to implied volatility. They were right on one count and wrong on the other—and in most cases, they lost money.
DS: In your seminars, you talk quite a bit about people looking for magical option positions.
MO: Yes, that’s a common mistake. A lot of people want to find a position that allows them to have their cake and eat it too. Among dealers or neutral options speculators, they tend to be addicted premium buyers or sellers. They might think, “If I’m only long options, then all the big moves will be in my favor and everything is going to be great.” Or if they’re premium sellers, they think, “If I’m only short options, the options will decay and this is the right position.” A speculative trader might say, “I want to find a profit opportunity that has no real risk to it or one where the risks don’t matter.”
|“When I’ve lost a lot of money trading options, it was never because of an incremental change in anything. It always happened because there was a big change in underlying price, time, implied volatility or some combination of those things.”
From a hedger’s point of view, it’s a little different. It usually means, “I would like to hedge my position, but I don’t want there to be any cost and I don’t want to give up my profit potential.” In the 1980s and early 1990s, the magical position for hedgers tended to be some kind of zero-premium position. Because hedgers don’t like to write checks for option positions, the magical position often turns out to be a collar or a participating forward or any other of a wide variety of zero-premium positions. Sometimes, these are called “zero cost” positions. Eventually, naive hedgers get an expensive lesson in the nature of opportunity costs.
But the fact of the matter is that there are no magical positions out there. Every position has some good and bad in it, and any position can be a good idea or a bad idea, depending on the circumstances.
Another way people look for options magic is by trying to use options to make the past untrue. I have often run into market-makers who have problems and want to talk about them. A common problem is, “I put on this position, and I know I shouldn’t have been that short. It started to go against me and I should have adjusted, but I didn’t—and now I am really in trouble.” What they really want to know is, How can I do an option trade to make none of this true anymore?
You see the same thing in corporate hedging. A corporate hedger has an asset or liability that he should have hedged but didn’t, or one he should have hedged more thoroughly. The trade then went against him, and now he is looking around for a way to make the past untrue.
DS: It’s just avoiding taking losses?
MO: In some cases that’s true, but I’ve heard people say, “It’s my responsibility to get this back.” That can be scary, because they might take a lot of risk to give themselves a chance to make everything OK.
DS: I think that’s what Nick Leeson said.
MO: I wouldn’t be surprised.
DS: What other common mistakes do you see in options trading?
MO: We often see people draw inappropriate conclusions from short-term results. Too many managers excessively punish or reward individual traders based on short-term P&Ls. But in the options business, like so many businesses, there are a lot of random ups and downs. If you are too quick to jump all over your traders when they lose money and too quick to turn them into heroes when they make money, your traders are going to think up positions that have high probabilities of winning. But in a nonlinear, multidimensional business, a position that has a high probability of winning may or may not be a good position.
DS: It may be exposing you to a huge tail risk.
MO: That’s one possibility. The problem could be a remote event that would be disastrous for your organization. Or the problem could be the average result. You could simply be exposing yourself to a position that might win most of the time, but will lose over the long term. It’s easy to think up positions that are going to win 70 percent, 80 percent or 90 percent of the time, only to find that the large losses inevitably eat up all those small wins.
DS: Could you give me an example?
MO: Back in the mid-1980s, most major commercial banks were getting into the foreign exchange and interest rate options business. To attract them as clients, brokers wanted to recommend a winning trade. Many recommended strangle-selling, in which you sell an out-of-the-money call and an out-of-the-money put—and very likely collect the premiums on both sides.
If the calls and the puts are priced richly enough, and if the underlying commodity isn’t likely to be too volatile, the trade might actually be a good idea. But if you sell the strangle too cheaply, ultimately the approach is going to be a loser. And if you do it in big size, you will ultimately get hurt, because there will be an occasional huge loss to go with all your small wins. But these kinds of trades can seem quite attractive, even if they’re not particularly good. If you’ve forgotten that remote events happen, they can even seem like options magic.
DS: So the head of a trading desk relying on short-term results is not measuring the risk his traders have taken on fairly or accurately enough.
MO: Exactly. He’s doing two things. First, he is measuring the quality of the trade simply by the frequency of profits, or by the likelihood of making profits—not by the average profit over time. Second, he is measuring the risk only in terms of everyday events and not in terms of remote events. Of course, in the 1990s, value-at-risk foolishness has lured a lot of organizations into feeling safe while ignoring remote risks.
DS: Of course, you can also make mistakes by ignoring short-term results.
MO: Sure. You hear a lot of rationalizations such as “This trade didn’t work out, but if I keep doing it, it will work out.” Sometimes that’s true, and sometimes not. You can’t ignore short-term results, but on the other hand, you can’t assume that short-term results validate a strategy.
|“In a nonlinear, multidimensional business, a position that has a high probability of winning may or may not be a good position.”
Within a large organization, a big part of the problem can be that managers don’t want to have to learn the options business. They don’t want to manage the process. They just want to manage the results. That’s not good enough. Short-term results don’t convey enough information. Managers need to be coaches.
DS: I don’t suppose the trap of focusing on short-term results is limited to brokers and managers.
MO: No. Traders do it all the time. Maybe they always have. In 1975, when I began thinking about becoming an options trader, I spent some time on the CBOE floor to see if I could learn something. It was largely a premium selling community. A lot of traders thought the idea was to sell calls that were a couple of strike prices out of the money and that had only a few weeks to go. They thought that during the last month before expiration, a stock wasn’t going to run through more than one strike price. When you asked them why they were so sure that was true, the answer would often be, “I tried it through three expirations in a row and made money every time.”
My old partner, Jim Piper, used to talk about “the permanent present tense.” That refers to the way traders use the present tense when the past tense or future tense would be more useful. For example, you will hear traders say, “Buying premium is what works now,” or “Selling premium is what works now.” What that means is, “This strategy has been working lately, so it’s what I’m doing for the future.” The present tense conveys a fantasy of stability and predictability. It also allows them to think it isn’t their fault if they lose money.
DS: What other weaknesses are traders likely to have?
MO: O’Connell & Piper used to have a client in Chicago—a smart market-maker—who was good at hiring traders. He said there are two things that are difficult to find in a trader. The most difficult thing to find is a trader who can take a big loss and not have it upset him to the point that he changes his behavior. The second most difficult thing to find is a trader whose perspective is not affected by an unusually large gain.
DS: It’s like in any business or undertaking. If you have a failure, you think that something needs to be changed. If you have a success, you think you know why. But that might not be the case.
MO: Right. What’s different in our business is that it all happens faster. If you make money building hotels, the same formula might work for years. Then, you might find it doesn’t work for a while, and it might be some time before it makes sense to build hotels again. If you’re smart or lucky, you’ll change your behavior at just the right time. If conditions don’t change too often, you won’t really know whether you were smart or lucky—or dumb or unlucky. But we deal in much shorter cycles, and that’s why our weaknesses are so visible in our behavior.
DS: Any other common options fallacies that come to mind?
MO: The fantasy of continuous markets. Often dealers and others will trade as if they will always have a chance to get out. An assumption of most models is that markets are going to be continuous. That’s usually true enough, but every once in a while markets are not continuous and people are severely hurt.
A related fantasy is excessive confidence in stop orders. Traders will put on negative-gamma positions. These are positions that get shorter as the underlying price goes up and get longer as the underlying price goes down. Traders will try to defend the position with stop orders. Of course, stop orders work until they don’t work anymore. Most of the time, stop orders get filled easily. But sometimes they don’t get done at anywhere near the price you expected. You see this often in foreign exchange, but also in commodities. People will leave overnight stop orders and wind up getting terrible fills because the markets were illiquid when their stop limit got hit.
DS: We’ve talked a lot about speculative traders, but you’ve also done a lot of work with corporate hedgers. Do you have any advice for them?
MO: Hedgers focus too much on expiration characteristics. Corporate hedgers or institutional investor hedgers draw expiration graphs and think they understand them, so they put on positions and let them sit until expiration. Of course, when you draw an expiration graph, one of the things you have to ask yourself is, How sure are you that you won’t touch this position? Because if there is a reasonable chance that you are going to touch the position before expiration, the expiration graph won’t matter because your P&L profile will look different.
You often see somebody draw an expiration graph and say, “This is great except for this one little piece of the graph over here where I’ve got a problem, but if it goes that way I’ll make an adjustment.” So once again, the whole expiration graph doesn’t mean much. Even in 1999, we still see hedgers looking at expiration graphs for positions with barrier options. They’re really kidding themselves.
|“You still hear people say, ‘We don’t like the kind of size one of our traders does, but what can we do about it? He’s making money.’”
A related problem is the tendency to leave a position untouched until expiration. An initial position in an option hedge is usually based on views on the direction or volatility of the underlying price or on changes in implied volatility. The nonlinear, multidimensional characteristics of the position should reflect these views as well as some considerations of risk tolerance. But over the life of the position, it’s almost certain that some of these views or characteristics are going to change. And if these things have changed, then the hedger probably doesn’t have the best hedge or even an appropriate hedge anymore. The simplicity and comfort of focusing on the expiration characteristics can be easy, but for many hedgers it’s time to move to a more sophisticated approach.
DS: There’s been a lot of talk about improving the risk management of traders in the past few years, but based on your comments we haven’t come too far.
MO: Our computers have advanced faster than human nature. Although most big trading firms these days have large, sophisticated risk management operations, you still hear people say, “We don’t like the kind of size one of our traders does, but what can we do about it? He’s making money.” You still see situations in which traders commit risk violations and the manager overlooks them because the position made money. Those managers have a lot of nerve cashing their paychecks.
|Was this information valuable?|
Subscribe to Derivatives Strategy by clicking here!