Thursday, July 15, 2010

Sweeney, Campaign Trading

While I wait for new books to arrive it’s often worthwhile to go back to some old ones. John Sweeney’s Campaign Trading: Tactics and Strategies to Exploit the Markets (Wiley, 1996) is a book I return to more frequently than most, even though it’s no page turner. Sweeney is best known for his development and analysis of the notion of maximum adverse excursion (MAE), both in this book and in another one (Maximum Adverse Excursion: Analyzing Price Fluctuations for Trading Management) published a year later. He also introduced, although he didn’t study it as rigorously, the idea of favorable excursions (MaxFE and MinFE). Since I personally keep some rough statistics on a less clearly measurable variation of these ideas, I like to touch base now and again with the real McCoy.

The book’s title promises far more than it delivers, but even the promise and the few words about it are enticing. Campaign trading means trading continuously across different markets, using a set of rules tailored to each market that can both identify the mode the market is in and allow traders to be active in all market phases—basing, breakout, trending, topping, and retreat. (A sidenote here: Although Sweeney glosses over the distinction, I don’t believe that the set of meta-rules that identify market regimes can be the same as the set of entry and exit rules. Can’t we ever get past using moving average crossovers as the default setting for every occasion?) Campaign trading also means using “loss-control techniques to generate profitability even when our trading signals aren’t God-given, and varying tactics to compound our position.” (pp. 23-24)

Sweeney is now on stride. MAE, the maximum intraday adverse price movement from the point of entry for a trading system, gives us an idea of whether the system is even tradable given our account size. If it is tradable, backtested trades should be divided into winners and losers, with the MAE tabulated for each. A decent set of trading rules should generate distinctly different distributions for good trades (that is, trades with a smaller MAE) and bad trades. Finally, traders should adjust their stops “to the combination of loss and profit level [they think] is appropriate.” (p. 56) (I should note in passing that Sweeney invokes the notion of stops in this book to keep things simple, though he suggests that in practice traders should consider using options rather than stops to minimize loss. His discussion of options toward the end of the book, however, is itself simplistic. Anyone contemplating replacing stops with options should do a lot more homework.)

What are the benefits of using this experientially-based system of setting stops? Sweeney answers: “What I really like about this sort of analysis . . . is that it’s imperceptible to competitive traders and analysts bound up in parametric statistics and charts. Moreover, because it’s applicable to any set of trading rules, its results are unique and my stops are less likely to rest around the levels where everyone else normally clusters.” (p. 61)

Sweeney is a frustrating writer because he throws out a lot of ideas that he doesn’t follow through on and belabors some obvious points. But every time I go back to this book I catch the glimmer of another bright idea. It’s seductive.

No comments:

Post a Comment