A Dozen Questions to Make You a Better Trader
(Guest Commentary by Bill Rempel – December 13 , 2007)
Dear Subscribers and Readers,
Before we go on to Bill Rempel's excellent guest commentary for this month, I would like to quickly point out the “divergence” we are witnessing in natural gas prices versus other energy prices. In light of this year's record natural gas storage levels, the continued lack of LNG import capacity (thus eliminating the opportunity to “arbitrage” between global natural gas prices and the domestic price at the Henry Hub in Louisiana), the lack of “fuel switching” among domestic industrial users (many of these left the country in light of the natural gas spikes we witnessed during the 2000/2001 and the 2005/2006 winters), and the continued tightness in the global crude oil market, the traditional “6-to-1” ratio between oil prices and the natural gas winter contracts have completely broken down. Instead, we are now witnessing a ratio more than twice that – of approximately 12.5 – basis the January 2008 futures contracts. However, even within the natural gas complex, we are also witnessing a divergence between domestic natural gas prices versus natural gas prices in other major regions. The following chart, courtesy of Goldman Sachs, shows the current divergence between NYMEX natural gas prices versus natural gas prices in other parts of the world – namely prices in the UK, Japan, and Continental Europe:
At some point, natural gas will be a “screaming buy.” But until I see some signs that the current US-centric credit crunch is easing, or until I see an even greater divergence, I would hold off on buying. Moreover, given that this winter is projected to be a warmer-than-average winter, this author would most likely purchase contracts or options for the 2009/2010 winter at the earliest, since storage levels will most likely remain ample going into the 2008/2009 winter. Make no mistake: At some point, the natural gas market will be too difficult to resist for the “hedgies” and momentum traders alike, but we are still not there yet.
Let us now go on to Bill's commentary for this month.
For those who had wanted to learn more about individual stocks, the art of stock selection, and model-based trading/investing, it is again time to see what one of our regular guest commentators, Bill Rempel, has to say. Bill is a prolific writing on the stock market and individual stocks and is the author of a very active market blog at: http://billrempel.com
In this commentary, Bill is going to discuss one of the most important requirements in investing or trading successfully – that of building a trading system that suits your temperament, long-term needs, and other requirements. Bill also argues that – contrary to what most believe – everyone has a trading system. The only difference is the degree of soundness of each system – ranging from a “system” based on pure market emotions to a purely complex quantitative system requiring the use of a supercomputer. Trust me, this commentary is a fascinating read. Without further ado, following is biography of Bill:
Bill Rempel (aka nodoodahs) is an active poster on the MarketThoughts forum as well as a few others around the web. Bill is a regular, monthly guest commentator on our website (see “Glass Half Full For Value Stocks” for his last guest commentary). Bill graduated from Caddo Magnet High School (a high school for nerds) back in 1985 and proceeded to learn the hard way when he drank his way out of a scholarship to Tulane later that year. After a few years of sweating for a living, he decided to go back to school, and graduated from LSU-Shreveport in 1995 with a Bachelors in Mathematics - all the while working the overnight shift stocking shelves in a grocery store.
Post-college, Bill has been in the P&C insurance industry as an actuary, product manager, and pricing manager. Bill and his wife Millie are amateur investors with a variety of holdings, but they prefer to buy and hold value investments. In typical "value" style, they live cheap, driving old cars and preferring to save or invest instead of buying fancy "stuff."
Disclaimer: This commentary is solely meant for education purposes and is not intended as investment advice. Please note that the opinions expressed in this commentary are those of the individual author and do not necessarily represent the opinion of MarketThoughts LLC or its management.
Over the last couple of years, one of my driving interests has been examining and classifying the different types of trading systems. My first attempt at classifying systems started with two tenets. The two tenets were that all traders were system traders, and that all systems answered five basic "nuts and bolts" style questions. I've come to realize that the basic questions about systems are far deeper, and more theoretical, than I touched on with those initial five questions. I've also come to realize that the answers to these questions are even more vitally important to the trader than I first thought they were. What follows is an expanded list of questions that, if you can answer them, will make you a much better trader in the long run.
(1) What level of activity am I comfortable with? If I can't watch the market constantly, because of work or travel commitments, or if I were a wealthy retiree, it would make no sense to use or develop a system that required three monitors and access to "Level II" quotes. I would have other, perhaps better, things to do with my time.
Activity is akin to leverage: an active good system will generate a higher annualized return than a semi-active system that is just as good, and vice versa. I think of activity as "inventory turnover" in a retail store, and the margin on items sold as "system profitability," or the day-traders' expectancy (a unit of average dollars gained per dollar risked). If I can make a nickel on every item, I should sell more! If I lose a nickel on every item, selling more just hastens my eventual exit from trading.
(2) How much individual discretion will be allowed in my trading? I've had many conversations with readers of my site (and MarketThoughts) about discretionary versus mechanical systems. Many of the "system edges" discussed in question (7) strike to the core of discretionary systems, because there are so many systems that capitalize on the foibles of human behavior. If my system(s) is (are) based on our own flawed human emotional responses (as a collective market), they can I consistently trust myself to respond to the market in the way the system prescribes? Or do I respond to my flawed nature by making non-discretionary system rules that I follow, thereby moving the discretion away from system execution in the heat of the battle, and towards the safety and comfort of system design in my computerized backtesting?
(3) What is the risk level that I am comfortable with in my trading? There are almost as many ways to define risk as there are traders and systems, but a good definition of risk is needed in order to proceed. Again, I look at this as a self-knowledge issue first, and then a question of design parameters. Retirees may be looking for consistency, with the real risk being the posting of any negative returns. People who are trading to build wealth may look at an occasional negative quarter or even year as just a momentary stumbling block, with the real risk being the failure to achieve a sufficiently high return over the next 20 years or more.
Institutions and funds have different risk parameters than individuals, and not necessarily in the way most people might think. Some funds may have a minimum acceptable return that isn't much above that of the risk-free return, but it IS above it, and that fund may have zero tolerance for drawdown. Meanwhile, a hedge fund may have targets that are harder to meet, and a tolerance for some risk, but if they experience a drawdown (reduction of equity or prolonged "losing streak") exceeding some percentage, say 30%, it's almost a given that investors will be pulling their money from the fund – regardless of how recoverable such a drawdown may be.
(4) What is the payout distribution of the system I'm trading? There probably is some technical, precise, industry term for what I am describing, but I'm not aware of it. I start by imagining a system that generates "sure bets," trades that have a very high probability of winning. In most cases, the wins generated by systems like this are small, and proportionally the same size as the losses. I think of scalping as an example, or maybe the 2-day RSI system or "Q Crash" system. It may be psychologically easier for some people to trade a system with lots of small wins, because the number of losing streaks, and their average duration, will be smaller.
One extreme payout distribution comes from long-term trend-following (LTTF) systems applied to commodities futures, and is somewhat similar to buying out of the money (OTM) calls. Every once in while comes a big, huge payout, but the majority of the trades are small losses with a few small wins. This is a very tough system for most humans to trade, since they have to deal with long losing streaks and potentially high drawdowns, and they can't miss a trade, because they don't know beforehand which trade might make their year.
At the other extreme is the OTM put seller, who generates consistent income from premiums received, and has a huge winning percentage, until and unless the market moves dramatically against them – with a very rare, but very large, loss in the distribution.
This issue is tangentially related to items (1) and (2), above. An extremely active system, say, 4 or more trades a day, will tend to smooth out the results, making the distribution of returns on a monthly basis a lot more compact than the distribution of returns on a per-trade basis. A system which generates only a couple of dozen trades a year, like "value investing" according to the "Magic Formula" in The Little Book That Beats The Market, will have a pretty wild monthly distribution of returns compared to it's more-active counterpart.
(5) Why am I trading? Is the goal income replacement, or growth of capital? Is it a system to be traded while living off the proceeds? Is this a system that replaces one portion of an asset allocation, i.e. something done instead of indexing to U.S. stocks, or is the system a method of asset allocation? Is the system merely a subsystem in a larger group of systems used?
(6) How much capital do I have to be deployed with (or through) my trading? This is of keen interest to not only the retail trader, but also to the institutional trader. Capital needs can be defined by the system, and the system design may, in turn, be defined by (or ruled out as unacceptable by) the capital requirement.
It's possible for me to put together some very attractive LTTF systems using a diversified basket of commodities futures contracts. The amount of long-term CAGR (compounded annualized growth rate) that is achievable, even in relation to some gi-normous drawdowns, is impressive. This is made possible through the relative non-correlation of the individual contracts and the inherent leverage in the contracts. Here's the catch, I would need well over a million dollars (preferably several million dollars) in capital to run through that system without having a significant risk of ruin. Even larger in scale, hyperactive computerized trading of spurious pair correlations can be accomplished on the scale that Goldman Sachs' (GS) hedge funds trade, just as highly leveraged interest rate spread trades can be done on their scale – but these things are likely out of reach for me, or anybody reading my work.
On the other end of the spectrum, a micro-cap momentum or "pink sheet" strategy simply can't be accomplished with too much capital. CAN-SLIM, in its original flavor, required stocks with floats of less than 20 million shares, which to some extent dictates a maximum capital requirement. Many strategies which work well on one scale (hold the top 20 qualifiers and minimum market cap of $100 million) don't generate quite the same Sharpe ratio when tested on another scale (hold the top 100 qualifiers and minimum market cap of $1 billion), and might therefore be discarded by one trader, and not by another.
(7) What is my trading "edge?" "Edge" is usually used to refer to something that a trader knows, or exploits, that other traders either don't know or can't exploit. Truly, this is a nebulous concept. I think that edges fall into two broad categories, structural and psychological, and that psychological edges have further subcategories.
Speculators have a structural "edge" in commodities futures because the entity on the other side of most trades isn't trying to make money from the trade. Huh? The producer who used silver or copper to manufacture a product didn't enter the commodities market to profit from trading silver or copper futures, they entered the market in order to smooth their expenses (and thereby smooth their earnings), and they don't really have a dog in the fight as to whether copper goes up or down. The return the speculators get (on average) is a premium paid by industrial participants in order to defray longer-term risk. The same holds true for producers and users of soft commodities, as well as producers of copper, silver, or gold, who enter the market not a directional bet, but to hedge their income swings. That hedging produces a net structural edge for the class of participants who bear the risk, i.e., the speculators. Of course, now that the futures market has been the playground of hedge funds for many years, this "edge" is diluted amongst many players, some more sophisticated than others.
Some structural edges come from being a third party to primary transactions, such as market makers. Others may come from trading in a timeframe where larger participants are less sensitive to transaction price, such as scalpers. The timeframe edge is one that bleeds over into some of the psychological edges, as well.
The psychological edges are probably what most people think about when discussing trading strategies, and are sometimes discussed in academic literature as "anomalies." Post-earnings announcement drift is an anomaly that could be said to have both a psychological edge and a structural edge. Large institutions buy slowly and leave time for nimble small traders to ride along. Or is it that most parties, including large institutions, are slow to recognize the implications of the announcement? It's a good question.
The science of "behavioral finance" attempts to classify the reasons for various psychological edges, but they fall into more recognizable categories once we talk about how to find these edges. There are two main ways to find them. We can examine the fundamental aspects underlying the market movements, or we can examine the market movements themselves. One of these is fundamental analysis, and the other is technical analysis.
Fundamental analysis could be as simple as "the Russian wheat harvest isn't going to be as bad as most people think," or as complicated as applying the "accrual anomaly" to the cash flow and income statements of hundreds of prospective companies to determine which stocks to buy. The "macro-econ-for-dummies" crowd is engaging in fundamental analysis, of a sort.
Technical analysis applies to any market where price and volume data are available, and it is a study of past sentiment used in order to determine what future sentiment may be. It can be as simple as chart pattern recognition, which is unfortunately not very quantifiable in terms of effectiveness for most practitioners. It can be as complicated and quantifiable as the hedge fund that trades 100's of times a day on the same pair of stocks, working on 100's of different pairs of stocks simultaneously, based on spurious correlations observed over the last 9 months of tick data. It could be both simple and quantifiable, as in the 2-day RSI system, the "Q Crash" system, or a three-moving-average crossover system applied to 20 different commodities futures contracts simultaneously.
When I combine the two, as a value investor does when he compares a calculated "intrinsic value" to a "market price," I practice what I call "FundaTechnical" analysis.
The extended discussion of "edges" and their types could fill many different articles, but the purpose of discussing the "edge" in terms of systems design is to make sure my system has one! I don't think it's enough to be able to backtest a system and come up with a positive expectancy or a solid return over time; I prefer to see that return paired up with a reason for that return, or in other words, an "edge." It can be as simple as "trends tend to persist over this timeframe" or "extremes of movement tend to be mean-reverting over this other timeframe," but it needs to exist.
Now, at last, I am back to the very "nuts and bolts" questions.
(8) What am I trading? Does the system trade stocks, options, bonds, funds, ETFs, futures, real estate, baseball cards, or some mix thereof? I'm currently tracking and trading a system that picks from 90 different ETFs, and tracking three other systems, two that trade stocks of companies on U.S. exchanges, and one that moves between only three issues (SPY, IEF, and cash). The more mechanical the system, the more likely that this is very strictly limited and non-discretionary. Many CTA's and long-term trend followers trade a basket of twelve to twenty of future contracts. There are really no limits to this answer, but it does help define the system, and the answers to previous questions help define the answer to this one.
(9) What brought this particular trade to my attention? In parlance, this is the set up, and it should be directly related to my system's "edge." Some set-ups may be as specific as “the contract price equals the high for the last twenty-one sessions” or they may be general, as in “the stock is trending with the market, is above its 10 period MA, and just printed a narrow-range candle.” It's important to note that “the stock is trading under net current asset value” is a set up for a trade, nothing more, nothing less. Generally speaking, one needs a longer time frame to capitalize on a NCAV set up then one needs to capitalize on a narrow-range candle setup, but they are both set ups, nonetheless.
(10) How much do I buy or sell? Multiple posts could be, and have been, done on the subject of position sizing and risk control. This question is the least thought-about and most discretionary of all the "nuts and bolts" questions, for most traders and investors.
(11) When do I enter a position? What many people think of as a complete system is usually just a set of entry rules. This is something that could be more or less automated depending on the system. The CTA with a 21-day high set up for a contract may start entry on the next tick up, the day trader may enter when a long bar violates the range of a narrow bar, the swing or position trader may make decisions after-hours and place orders for the market open. There could be multiple entries, for example, pyramiding shorts to maintain constant capital allocation, or the Original Turtles rules for adding to an open position. In some systems, this could be a time rule, along with (12), something on the order of "rebalance every X months."
(12) When do I exit a position? This is probably the second least thought-about of the "nuts and bolts" questions, other than position size. Exit rules run the gamut from automated 13-day breakdowns, ATR / volatility-based rules, contract expiration, Fibonacci retracements or projections, and, of course, the ever-popular “I'm scared, Mommie!” exit rule.
Every trader is a system trader, whether he or she thinks so, or not. We generally walk around blind to the "systems" we follow in our lives, and are only dimly aware of them as "habits" – and we call them "bad habits" if they are suboptimal systems. The trader who hasn't examined their own behavior in the context of the dozen questions above is still following a system, albeit one that they're probably not consciously aware of, and one that is possibly far less than optimal.
Answering the above dozen questions, and thinking about your trading in terms of system design, even if it's a discretionary system, is probably one of the best things you can do to take your trading to the next level.