Sunday, March 05, 2006

Buy Premium! (Part II)

In the first part of the article I talked about the virtues of buying options and expanded on employing volatility view as the basis for a long strategy.

While vol trading is a good reason to get long options, there are a few drawbacks:
  1. Often you have to continuously flip stock against your long gamma. This creates non-negligible transaction costs.
  2. Every time you flip the stock you have to make a decision to limit your profit running in one direction. That's a tough thing to do, time and again.
  3. You often have to keep your positions on for many weeks so that your stock profits outrun the time decay of your long options. So, besides constantly having to trade around your position (that takes focus away from other things), you also have to commit your capital for extended periods.

In spite of these drawbacks, I still do vol trading all the time, but more often than not I look to buy options in order to capitalize on a highly probable directional outcome. Before I go into this let me state for the record that directional trading is a very tough game. In fact I think it is by far tougher than any other game in town.

Why is it tough?

  • It's the most level playing field for traders. Everyone has pretty much the same level of access and thus, unless you are a specialist or an institution filling a massive order, there is no inherent edge available to anyone.
  • Directional trading has been around for much longer than any other type. Thus every imaginable method has been developed and tried in every market that goes up and down. You have to assume that they were developed by a lot of very smart people who have tried very hard and then failed miserably. Notably, a few winners do stand alone. Very alone.
  • The data needed to run strategies and evaluate new directional methods is readily available. The facilities needed to grind this data are equally ubiquitous. Again, no mystery there.
  • There is more randomness in the evident directional moves that most traders choose to believe. That's their choice but the reality is that most of the time the markets are chaotic and have very low odds of showing a particular direction (the one that agrees with yours)

Other factors, too numerable to be listed, exist. The point is, if you're going to play this game, you'll need every imaginable advantage you can find.

My advantage is using options, and not the underlying, for directional positions. There are a lot of reasons for this but the two most important ones are below:

  1. If you're long an option and you're wrong about the direction, you lose money at an ever slowing pace. This is due to delta curvature (gamma) where as the underlying moves away from you your delta exposure automatically drops off. So, if you started with an ATM call with 50 delta, you end up with 25 delta after a couple of points against you.
  2. You can afford to be wrong. This is basically an elaboration of (1). Let's say you want to buy 10 calls. You know that being right at the time of purchase is unlikely. So, you buy 3 instead of 10. Then as you end up wrong, you buy more. Yes, you cost average. And here is why it's not an evil thing to do: you buy options at lower cost and with lower delta exposure. This last fact is crucial. The more and lower you buy, the slower is your rate of loss. And as your rate of losing slows down, your odds of making a profit on a reversal are increasing.

Caveat emptor: I have completely factored out time decay. I chose to discount it because I generally buy options with at least 10 weeks to expiration and my directional setups have an expectation of working (or not) within 5-7 trading days. Thus, theta is minimal.

Other reasons exist but, contrary to popular belief, are not nearly as key: the most frequently cited ones are limited risk and greater leverage.

Now, if you consider (1) and (2) carefully it is obvious that cost averaging should be done close to turning points of the underlying. Naturally, those points exist at substanstially extreme levels. Without giving too much away, let's quantify "substantially extreme":

  1. Two or more standard deviations away from some average or mean.
  2. Maximally short in terms of elapsed time.

These two points are not to be taken literally. They comprise the basis for the directional model that has a respectable level of probabilistic outcome. The actual parameters and numbers with respect to the magnitude of the move and the time factor depend on the underlying, the sector, the market and, ultimately, the trader.

Cheers,
Dmitry

0 Comments:

Post a Comment

<< Home