Article Reviews, Trading

Quotable Quotes from William Eckhardt (Mechanical Trend-Following Systems Trading)

I was reading more about William Eckhardt (or the Turtles trading experiment fame with Richard Dennis).

Eckhardt launched his own commodity trading advisor (CTA), Eckhardt Trading Company, in 1991, which has produced a compound annual return of 17.35% over 20 years and earned 21.09% in 2010. There are some updated statistics on this website, which showed these results over the period July 1991 to May 2013:

  • Annualized return of 14.5% vs. 9.2% for S&P
  • Sharpe ratio of 0.6 vs. 0.45 for S&P
  • Worst month of -18.3% (1994) vs. -16.8% for S&P (2008)
  • Worst drawdown of -27.11% (1992) vs. -50.95% for S&P (GFC)
  • Worst year was -15.64% in 2011
  • Best year was 57.95% in 1993
  • Positive returns for 17 out of 23 years (~74%)
  • Positive returns for 153 out of 262 months (~58%)

From what I’ve read, he is really a hard-core brilliant scientist in systems trading. He is the crème de la crème, standing at the top of the world of mechanical trend-following trading.

That’s why his results are particularly noteworthy and sobering. If the top guy is doing a 14.5% annualized return, understandably tied to the risk exposure taken, it really humbles my return expectations for my own trading based on trend-following.

I have categorized key points below into three categories: (1) psychology, (2) methodology, and (3) systems development. There are a lot of very good points shared by William Eckhardt on many aspects of trading and systems development. I highly encourage readers to read the entire post.


You Cannot Bring Normal Human Tendencies To Trading

  • If a betting game among a certain number of participants is played long enough, eventually one player will have all the money. If there is any skill involved, it will accelerate the process of concentrating all the stakes in a few hands. Something like this happens in the market.
  • There is a persistent overall tendency for equity to flow from the many to the few. In the long run, the majority loses.
  • The implication for the trader is that to win you have to act like the minority. If you bring normal human habits and tendencies to trading, you’ll gravitate toward the majority and inevitably lose.
  • Anyone with average intelligence can learn to trade. This is not rocket science. However, it’s much easier to learn what you should do in trading than to do it. Good systems tend to violate normal human tendencies. Of the people who can learn the basics, only a small percentage will be successful traders.
  • If you’re playing for emotional satisfaction, you’re bound to lose, because what feels good is often the wrong thing to do. Richard Dennis used to say, somewhat facetiously, “If it feels good, don’t do it.” In fact, one rule we taught the Turtles was: When all the criteria are in balance, do the thing you least want to do. You have to decide early on whether you’re playing for the fun or for the success. Whether you measure it in money or in some other way, to win at trading you have to be playing for the success.

Human Nature Does Not Operate to Maximize Gain, But the Chance of Gain

  • One common adage on this subject that is completely wrongheaded is: you can’t go broke taking profits. That’s precisely how many traders do go broke. While amateurs go broke by taking large losses, professionals go broke by taking small profits.
  • The problem in a nutshell is that human nature does not operate to maximize gain but rather to maximize the chance of gain.
  • The desire to maximize the number of winning trades (or minimize the number of losing trades) works against the trader. The success rate of trades is the least important performance statistic and may even be inversely related to performance.
  • Two of the cardinal sins of trading – giving losses too much rope and taking profits prematurely – are both attempts to make current positions more likely to succeed, to the severe detriment of long-term performance.

Emotional Makeup is More Important Than Intelligence

  • I haven’t seen much correlation between good trading and intelligence. Some outstanding traders are quite intelligent, but a few aren’t. Many outstanding intelligent people are horrible traders. Average intelligence is enough. Beyond that, emotional makeup is more important.
  • As I recall more than half the course revolved around developing the right attitude, guarding against debilitating emotions, how to think about risk, and how to handle success and failure.
  • Teaching the turtle system itself doesn’t take very long. I was saying you need less than 12 degrees of freedom in a system; versions of the turtle system had three or four.
  • We spent a lot of time talking about our theories on how to control risk; that was actually the bulk of the course. Attitude, emotional control, discipline; those things are harder to teach. All the turtles learned the system and learned the strategy; that was the easy part, but some of them brought the right attitude and right mental set to it and they prospered and became very rich. Others had a more halting career and did not succeed as well. They had the same training, but maybe they did not have the same emotional make-up.

If You Can’t Change Your Behavior, Be a Systems Trader

  • If a trader doesn’t know why he’s losing, then it’s hopeless unless he can find out what he’s doing wrong.
  • In the case of the trader who knows what he’s doing wrong, my advice is deceptively simple: He should stop doing what he is doing wrong. If he can’t change his behavior, this type of person should consider becoming a dogmatic system trader.

You Are Finished If You Keep Missing Good Trades

  • I take the point of view that missing an important trade is a much more serious error than making a bad trade.
  • If you make a bad trade and you have money management you are really not in much trouble. However, if you miss a good trade there is nowhere to turn. If you miss good trades with any regularity you’re finished.
  • For example, let’s say the market moves rapidly through your buying zone and you miss it, you miss your buy signal and instead wait for a retracement to maybe buy cheaper. But, the market just keeps going higher and higher and never retraces. Now what do you do? There’s a great temptation to reason that now it’s too high to buy. If you buy it now you’ll have an initiation price that’s too high? No, the initiation price simply won’t have the kind of significance you suppose it will have after the trade is made. You can’t miss these trades.
  • If you miss a good trade, you have nothing to protect you-that is, nothing in the system will assure that you eventually get in. Also, missing a good trade can be demoralizing and destabilizing, especially if you’ve been in the midst of a losing period. And like so many bad trading decisions, it ends up costing you more than just the money lost or not made on the trade. Missing a major trade tends to have a reverberating effect throughout your whole trading strategy. Sometimes it can be weeks before you get back on track.
  • Trading systems force discipline to make sure these trades are not missed.

Do Not Override Your System During the Trading Day

  • You should try to express your enthusiasm and ingenuity by doing research at night, not by overriding your system during the day.
  • Overriding is something you should do only in unexpected circumstances – and then only with great forethought. If you find yourself overriding routinely, it’s a sure sign that there’s something that you want in the system that hasn’t been included.
  • You can be creative in research but don’t trade creatively; in other words, stick to your systems.
  • If your trading system is inadequate, you shouldn’t use it. If your system is good, then stick to it faithfully.
  • In the meantime search vigorously for improvement. When the new system is ready you can change to it – you are not thereby failing to stick to your system. So there need be no conflict between persistence and change.

Losses Must Hurt But Avoid Snowball Scenarios

  • The people who survive avoid snowball scenarios in which bad trades cause them to become emotionally destabilized and make more bad trades. They are also able to feel the pain of losing. If you don’t feel the pain of a loss, then you’re in the same position as those unfortunate people who have no pain sensors. If they leave their hand on a hot stove, it will burn off. There is no way to survive in the world without pain. Similarly, in the markets, if the losses don’t hurt, your financial survival is tenuous.
  • I know of a few multimillionaires who started trading with inherited wealth. In each case, they lost it all because they didn’t feel the pain when they were losing. In those formative first few years of trading, they felt they could afford to lose. You’re much better off going into the market on a shoestring, feeling that you can’t afford to lose. I’d rather bet on somebody starting out with a few thousand dollars than on somebody who came in with millions.

Be Cautious After a Period of Wins

  • In many ways, large profits are even more insidious than large losses in terms of emotional destabilization. I think it’s important not to be emotionally attached to large profits. I’ve certainly made some of my worst trades after long periods of winning.
  • When you’re on a big winning streak, there’s a temptation to think that you’re doing something special, which will allow you to continue to propel yourself upward. You start to think that you can afford to make shoddy decisions. You can imagine what happens next. As a general rule, losses make you strong and profits make you weak.
  • It is a common notion that after you have profits from your original equity, you can start taking even greater risks because now you are playing with ‘their money’. We are sure you have heard this. Once you have profit, you’re playing with ‘their money’. It’s a comforting thought. It certainly can’t be as bad to lose ‘their money’ as ‘yours’? Right? Wrong. Why should it matter whom the money used to belong to? What matters is who it belongs to now and what to do about it. And in this case it all belongs to you.

The Market Tricks You to Trade Poorly

  • The market behaves much like an opponent who is trying to teach you to trade poorly.
  • Since most small to moderate profits tend to vanish, the market teaches you to cash them in before they get away.
  • Since the market spends more time in consolidations than in trends, it teaches you to buy dips and sell rallies.
  • Since the market trades through the same prices again and again and seems, if only you wait long
    enough, to return to prices it has visited before, it teaches you to hold on to bad trades.
  • The market likes to lull you into the false security of high success rate techniques, which often lose disastrously in the long run.
  • The general idea is that what works most of the time is nearly the opposite of what works in the long run.


Focus on Planning What to Do When the Bad Scenarios Happen

  • Don’t think about what the market’s going to do; you have absolutely no control over that. Think about what you’re going to do if it gets there.
  • In particular, you should spend no time at all thinking about those rosy scenarios in which the market goes your way, since in those situations, there’s nothing more for you to do.
  • Focus instead on those things you want least to happen and on what your response will be.

Entry Price Has No Bearing on Subsequent Trade Management

  • For example, let’s say the market moves rapidly through your buying zone and you miss it, you miss your buy signal and instead wait for a retracement to maybe buy cheaper. But, the market just keeps going higher and higher and never retraces. Now what do you do? There’s a great temptation to reason that now it’s too high to buy. If you buy it now you’ll have an initiation price that’s too high? No, the initiation price simply won’t have the kind of significance you suppose it will have after the trade is made. You can’t miss these trades.
  • Suppose two traders, A and B, who are alike in most respects except the amount of money they have. Suppose A has 10% less money but he initiates a trade first. He gets in earlier than B. By the time B puts the trade on, the two traders have exactly the same equity.
  • The best course of action has to be the same for both of these traders now. Mind you, these traders have very different entry prices. What this means is that once an initiation is made, it does not matter at all for subsequent decisions what the entry price was. It does not matter. Once you have made an initiation, what your initiation price was has no relevance. The trader must literally trade as though he doesn’t know what his initiation price is.

Entering on Retracement is Tricky

  • I don’t like to buy retracements. If the market is going up and I think I should be long, I’d rather buy when the market is strong than wait for a retracement.
  • Buying on a retracement is psychologically seductive because you feel you’re getting a bargain versus the price you saw a while ago. However, I feel that approach contains more than a drop of poison.
  • If the market has retraced enough to make a significant difference to your purchase price, then the trade is not nearly as good as it once was. Although the trade may still work, there’s an enhanced chance that the trend is turning.
  • Perhaps even more critical, a strategy of trying to buy on retracements will often result in your missing the trade entirely or being forced to buy at an even higher price. Buying on retracements is one of those ploys that gives psychological satisfaction rather than providing any benefits in terms of increased profits.
  • As a general rule, avoid those things that give you comfort; it’s usually false comfort.

Exits Are Much More Important Than Entries

  • Many systematic traders spend the majority of their time searching for good places to initiate. It just seems to be part of human nature to focus on the most hopeful point of the trading cycle.
  • Our research indicated that liquidations are vastly more important than initiations. If you initiate purely randomly, you do surprisingly well with a good liquidation criterion. In contrast, random liquidations will kill the best system.

Position Sizing Is More Important Than Entries

  • When and where you initiate a trade is a lot less important than how large you trade and how you liquidate.
  • Unfortunately, traders tend to put a great effort into trade initiation and let risk management stagnate.

Risk No More Than 2% Per Trade

  • You shouldn’t plan to risk more than 2 percent on a trade. Although, of course, you could still lose more if the market gaps beyond your intended point of exit.

Focus on the Price Action

  • An important feature of our approach is that we work almost exclusively with price, past and current.
  • One reason for this is that to make any progress in the early stages of quantitative investigation you usually have to reduce the relevant factors to one or two crucial variables. Price is definitely the variable traders live and die by, so it is the obvious candidate for investigation.
  • The other reason is that in a system that’s making good use of price information, it is very difficult to add other information without degradation. Pure price systems are close enough to the North Pole that any departure tends to bring you farther south.
  • We’ve been doing this for a long time, so at this point, most things we test degrade our system. But every year or two, we’ll find something that actually improves it.
  • A price chart is an attempt to model relevant aspects of price change. Price change is not linear displacement, whether vertical, horizontal or oblique. Nonetheless, price change can be represented as vertical displacement and time elapsed as horizontal displacement. Such a model, however, invariably supports relationships that does not correspond to anything in the original process.
  • The angular inclination of a trend on a price chart is a visually striking feature of this representation. Such angles have no intrinsic meaning for the price series, but this is one of the many factors (along with our facility for pattern recognition and wishful thinking) that contributes to our interpreting more from price charts than rigorous testing reveals is there.

Look at the Detailed Structural Information in Price Data and Not Just Summary Results

  • Our aversion to summary statistics that obliterate structure extends to the trading systems themselves. For instance, we avoid moving averages of price in making trades. Such moving averages are popular mostly because they’re mathematically tractable, but they smooth away all the structural information inherent in the price data.
  • Another popular tool, the price breakout, may be far better than the moving average, but it still eliminates most of the relevant structure. A breakout trader keeps two pieces of structural information, the high and the low for a given time period, but ignores all the price structure in between. For this and for other reasons we judiciously avoid breakout trading in all parts of all our systems.

Have an Erraticness Filter

  • In our case specifically, we have an erraticness filter which is influenced by volatility. Erraticness incorporates different measures of market spread. If market erraticness rises above a certain threshold, new trades in that market are blocked.
  • This is designed to eliminate a subset of potential trades that we think will add to the portfolios’ volatility without contributing much to their returns.
  • We introduced the erraticness filter near the end of March 1996, and it turned out to be very beneficial. For the first few years after we implemented the erraticness filter, in examining our Sharpe ratio the numerator got bigger and the denominator got smaller simultaneously. In my experience, that’s very rare.
  • A few months before things really began to fall apart [in 2008], our systems essentially shut down. They judged the market to be too erratic. When the crisis hit, we had small positions.
  • What we tend to do is just assess the fact that everything’s become riskier and more volatile, and we liquidate a proportion of everything.

Holding Periods

  • We have three packages which consist of 19 systems in all.
  • The short-term package has an average trading length of about 6 days; the medium term package has an average of about 12 days. The long-term package is over 60 days.
  • All of the systems trade independently and are designed to be profitable on their own.

Win Rate

  • Looking back, about one third of the trades have been winners, and two thirds losers. That’s been true for a long time.
  • The idea is you win in only a modest percentage of trades but you make these wins count.

Why Trend Following Works

  • I would offer a few reasons, all based on human nature. The first is that we’re not very good as a species at reasoning about probabilities. We are good at other things, such as estimating speed and distance. Take the ability to catch a baseball, for example; physicists tell us this requires integrating differential equations, which is of course quite complex. By comparison, we make mistakes in easy probability problems.
  • One consequence is that we tend to have only two responses to extremely small probabilities, neither of which is helpful: we ignore them completely or we exaggerate them. I would give the Anthrax scare some years back as an example of the latter. The probability that any single person would be infected with Anthrax was incredibly small, yet a lot of people were in hysterics.
  • The more typical response is to ignore very small probabilities altogether—to assume that they’re essentially zero. Let’s say there’s a one percent probability that beans were going up a dollar. That should make beans go up a penny. In fact, what would typically happen is that market participants would ignore that small probability, and the price wouldn’t respond at all. Let’s say the probability slowly creeps higher. At some point it registers on people’s mental scopes, so to speak. Then they tend to respond discontinuously to this continuous development.
  • Another example of how people behave unreasonably when faced with probabilities is the way they respond to lotteries. If you offer subjects a sure win and you offer them a lottery that’s a little better, they’ll take the sure win. On the other hand, if you offer them a sure loss or a lottery that’s a little worse but has a chance of recouping, they’ll take the lottery. Traders tend to follow the same—they take profits and they play with losses. This bias generates trends. The trendiness of prices seems to be grounded in human nature.Looking back, about one third of the trades have been winners, and two thirds losers. That’s been true for a long time.

Convert Ignorance to Profitability

  • I would characterize our overall approach as “conservative”. This does not mean that we avoid market risk, for market risk is the raw material from which profit is fashioned, but we are conservative about what we know and about what can be done.
  • My experience with Decision Theory indicates that knowing what it is you are ignorant of is in fact a powerful position to be in. The task of the trader is to locate those few areas where ignorance is not complete and to convert this information into profitability in an efficient way. False knowledge can be very detrimental to this process, but acknowledged ignorance can be quite beneficial.

Overbought / Oversold Indicators (RSI, Stochastics) Don’t Work

  • They’re close to zero in terms of their profit expectations. What these patterns make during market
    consolidations, they lose during trends.
  • For one thing, when you look at these indicators superimposed on a price chart, they look much better than they really are. The human eye tends to pick up the times these indicators accurately called minor tops and bottoms, but it misses all the false signals and the extent to which they were wrong during trends.
  • Formally, the mistake is the confusion between prior and posterior probabilities. For example, it’s true that a lot of extremes have reversal days. All that’s telling you is the probability of having a reversal day given a price extreme. What you really want to know is what the probability is. of having an extreme-that is, a sustained change in market trend-given that you have a reversal day. That is a very different probability. Just because one probability is high, it in no way implies that the other one is high as well. If 85 percent of all tops and bottoms have property X, but property X also occurs often enough in other places, using that indicator as a signal will rip you to shreds.

Cycle Analysis Don’t Work

  • There are very powerful scientific methods of cyclical analysis, particularly Fourier analysis. Fourier analysis has been tried again and again on market prices, starting in the late nineteenth century with the work of the French mathematician Louis Bachelier. All this scientific research has failed to uncover any systematic cyclic components in price data. This failure argues strongly against the validity of various trading systems based on cycles. And, I want to stress that the techniques for finding cycles are much stronger than the techniques for finding trends. Finding cycles is a classic scientific problem.
  • If you allow cycle periods to shrink and expand, skip beats, and even invert-as many of these cycle theorists (or, perhaps more accurately, cycle cranks) do-then you can fit cycles onto any data series that fluctuates. The bottom line is that rigorous statistical techniques, such as Fourier analysis, demonstrate that these alleged cycles are practically random.


Development Process

  • There are two parts to the process.
    • The first part is to develop a coherent portfolio theory: how to structure your trades, how to manage risk, etc. That truly is a scientific project in which you’re trying to develop things from first principles.
    • The second part is brainstorming for new trading ideas. It usually takes 70 to 100 false starts before we get something that we can use.
  • We pay a lot of attention to the foundations of the subject, to the soundness of our methodology, and to the correctness of our statistics.
  • In terms of the foundations of the subject, we rely heavily on Decision Theory and Utility Theory.

Prediction Models Do Not Help in Trading Systems

  • Statistical estimators probe particular features of the price series; they are equipped with confidence levels, give information about possible models, and are useful for prediction.
  • From the point of view of the modeler, trading systems do not locate specific features of the price series; they have no confidence levels and are useless for prediction. Worst of all, they say little about any possible model.
  • Trading systems can be highly remunerative, but they don’t tell the modeler what he or she needs to know. In the same way models, although valuable in other respects, do not help in designing trading systems.
  • Now superficially it seems like trading is a form of prediction but it really isn’t. If you design your system where you are trying to predict the market, then it doesn’t work. You have to concentrate on projecting losses, risk management and finding something that works, but if you are directly looking for prediction that tends to be self-stultifying.
  • And if you look at me as a predictor instead of as a trader — as a trader I am way ahead, as a predictor I am scoring about 35%, so I am not very good as a predictor. Those are different skills. But still even with trend followers you will hear people say, “Where do you think the market is going?” It is just human nature to try and approach this in terms of making a prediction.

Continuously Improve Your Trading Systems

  • Improve your trading or it will degrade; there’s no coasting in this game.
  • When I first began trading solely on the basis of price and was much more concerned than I should have been about the academic orthodoxy that futures market price change was pure white noise–a random walk–I made the following notebook entry: “How can the aggregate of traders and users arbitrage out a potentially unlimited number of nonlinear relationships?” The implication was that they could not.
  • Twenty-five years later, I am less confident about the continuing correctness of this answer. What I failed to take into consideration was the staggering explosion in information processing. This will only continue. Eventually artificial intelligence devices, superior to any human researcher, will effectively uncover all exploitable nonlinear relationships of price to price. Such relationships will be mined until technical analysis is no longer profitable. There is an irony in that dogmatic” random walk” theorists, dead wrong for a century, will turn out to have been prescient–futures markets will have been driven to randomness. The process has already begun.
  • I feel these developments are nearly assured (assuming no disruption of civilization). What is less clear is whether this will happen as rapidly as I predict–in 10 to 20 years. In the meantime, profitable trading will only get harder as increasingly more astute traders pursue progressively weaker statistical regularities. This is why it is necessary for a CTA continually to improve just to hold his or her own. The only consolation I can offer is that there are profits to be made participating in this process of randomization.

Limit Degrees of Freedom to 12, and Test Over a Large Sample Size

  • What most people use to ward it off is the in-sample/out-of-sample technique where they keep half their data for optimization and half their data for testing. That is an industry standard. We don’t do that; it wastes half of the data.
  • Now the two numbers that most determine if you are over-fitting are the number of degrees of freedom in the system. Every time you need a number to define the system, like a certain number of days back, a certain distance in price, a certain threshold, anything like that is a degree of freedom. The more degrees of freedom that you have the more likely that you are to over-fit. Now the other side of it is the number of trades you have. The more trades you have, the less you tend to over-fit, so you can afford slightly more degrees of freedom. We don’t allow more than 12 degrees of freedom in any system. If you put more bells and whistles on your system it is easy to get 40 degrees of freedom but we hold it to 12.
  • Seven or eight [degrees of freedom] is probably too many. Three or four is fine.
  • On the other side of that, for us to make a trade we have to have a sample of at least 1,800; we won’t make a trade unless we have 1,800 examples. That is our absolute minimum. Typically we would have 15,000 trades of a certain kind before we would make an inference as to whether we want to do it.
  • The reason you need so many is the heavy tail phenomena. It is not only that heavy tails cause extreme events, which can mess up your life, the real problem with the heavy tails is that they can weaken your ability to make proper inferences. Normal distribution people say that large samples kick in around 35. In other words, if you have a normal distribution and you are trying to estimate a mean, if you have more than 35 you’ve got a good estimate.
  • In contrast, with the kind of distributions we have with futures trading you can have hundreds of samples and they could still be inadequate; that is why we go for 1,800 as a minimum. That is strictly a function of the fatness of tails of the distribution.
  • You have to use robust statistical techniques and these robust statistical techniques are blunt instruments. They are data hogs, so both seem to be disadvantages but they have the advantages of tending to be correct.

Beware of Overfitting and Hidden / Bad Degrees of Freedom

  • There can also be hidden degrees of freedom. One can have structures within the system that can take on various alternative forms. If various alternatives are tested, it gives the system another chance to conform to past idiosyncrasies in the data.
  • Not only is it perilous to have too many degrees of freedom in your system, there are also “bad” degrees of freedom. Suppose a certain degree of freedom in your system impinges only on a very few oversized trends in me data and otherwise does not affect how the system trades. By affixing to accidental features of the small sample of large trends, such a degree of freedom can substantially contribute to overfitting, even though the overall number of degrees of freedom is manageable.

Take Care of the Tail Risk

  • The large-tail phenomenon means that most statistical tests overestimate reliability and underestimate risk. I don’t know if it’s possible to take advantage of this, but it’s important to protect yourself from it.
  • Tail risk is hard to estimate but we spent over 25 years on this project. We have worked on it really hard and we do have various techniques to deal with the fact that the tails are so heavy. It is absolutely crucial because the tail risk changes everything that we do. Every single part of designing and implementing the system is affected by the fact that you have more extreme values than you expect under any kind of normal model.
  • I have a little bit of trouble with the idea that the tail risk in futures trading is what is helping because I see it strictly as a hindrance, strictly as a problem to be overcome. I guess it helps to have these really big outsized moves. It is only going to help you if you treat it like a wild tiger. Trend-following doesn’t work only because of the tail risk but tail risk turns up the volume.

Trade Sizing Depends on Risk Aversion and Volatility

  • Risk aversion
    • When I was a young man I wanted to devise objective risk systems. In other words, once you have a system, what is the right size to trade, period.
    • After years of working on this I convinced myself that it did not have a unique answer. You need at least one subjective piece of the puzzle to put it together, and that is an individual’s risk aversion. Now that is subjective.
    • There is no rule that says how averse you should be to risk, that is an integral element of your personality. But unless you know how averse to risk you are or unless you can impute risk aversion to your clients, you really can’t settle the question of how big you should trade.
  • Volatility
    • Estimating volatility determines to a large extent what your position sizes should be.
    • A slight improvement in our volatility estimators can potentially produce a significant long-term benefit.

Don’t Set Your Trading Size at the Optimized Setting

  • On the subject of bet size, if you plot performance against position size, you get a graph that resembles
    one of those rightward-facing, high-foreheaded cartoon whales. The left side of the graph, corresponding to relatively small position size, is nearly linear; in this range an increase in trading size yields a proportionate increase in performance. But as you increase size beyond this range, the upward slope flattens out; this is because increasingly large drawdowns, which force you to trade smaller, inhibit your ability to come back after strings of losses. The theoretical optimum is reached right about where the whale’s blowhole would be. To the right of this optimum, the graph plummets; an average position size only modestly larger than the
    theoretical optimum gives a negative performance
  • Trading size is one aspect you don’t want to optimize. The optimum comes just before the precipice. Instead, your trading size should lie at the high end of the range in which the graph is still nearly straight.

Risk Management Needs to be Developed Together With Your Trading System, Not After

  • When and where you initiate a trade is a lot less important than how large you trade and how you liquidate. Unfortunately, traders tend to put a great effort into trade initiation and let risk management stagnate.
  • Small improvements in risk or volatility assessment may not be exciting, but they are among the most lasting and beneficial changes. One approach to avoid is to design the system first, then to tack on risk management.
  • System and risk management should be developed together; the connection should be seamless

Apply Utility Theory Incorporate Risk Aversion into Risk Management 

  • Our risk management techniques are based on utility theory. They take into account the fact that each dollar you make is a little smaller than the last one, and each dollar you lose is a little bigger than the last one. They allow you to quantify your own aversion to risk, and then to maximize expectations based on your risk aversion.
  • The objective of any investment is to achieve the highest returns based on your own risk tolerance, or in the case of a professional manager, the risk tolerance of your clients.
  • Note that there are two respects in which profits and losses are not equivalent. One is objective and has to do with nonlinearity. For example, it requires a 100% profit to balance a 50% loss. The second is subjective and has to do with risk aversion, for many people even the prospect of a 150% profit does not compensate for the risk of a 50% loss. Through Utility Theory, such imbalances can be treated in a rigorous, quantitative manner and in this way uniform and unified procedures can be developed.
  • We use only bounded utility functions in our work on risk management. The particular utility functions we
    use also have the desirable technical characteristic of optimal investment fractions being independent of
    absolute wealth level.

Take a Portfolio Approach to Risk Management 

  • Look at the question of risk management. Any trader who survives any length of time knows something about his subject, but in my experience, traders simply graft risk control on top of whatever else they are doing, often in an arbitrary way.
  • For instance, many prospective clients have asked me what’s the most I’ll lose on one trade. I can look up these statistics, but this is not something I would ordinarily pay any attention to. It doesn’t matter how little you lose on an individual trade, but how much you might lose on your whole portfolio.
  • You’re not going to keep a ship afloat just by making sure the leaks are small. The important thing is to limit portfolio risk, the trades will take care of themselves.

System Should Maximize Expected Utility

  • We have devised a portfolio theory quite different from the classical theory that permits factors such as risk aversion, the nonlinear imbalances between profits and drawdowns, and long-term utility growth to be built in at the ground floor.
  • They are all part of the formulas that define what it means for a system to be good. In this way, on even the most preliminary test run of a new idea we are forced to take into consideration the subtle and complex relations between drawdowns and long-term growth.
  • At ETC we are dedicated utility maximizers and pay particular attention to the rate of expected utility growth.

Evolutionary Algorithms for Optimization Process

  • We use evolutionary algorithms that we’ve developed in-house. To give you an idea of what that means, let’s say you have a system with certain parameters. Certain price points that you’re looking to hit, certain thresholds based on patterns you’ve observed. You can express a particular form of this system as a sequence of numbers, and treat that sequence exactly like a genome (a string of genes).
  • In order to test the system, you can run it with a given set of numbers. This will give you hypothetical performance figures which are analogous to an organism’s fitness. Then, just as in natural selection, you can cause genes to mutate or you can genetically recombine two genomes, always favoring those with higher fitness. The fitness can then “evolve” to be higher.
  • The objective is to find ways to identify trends within the massive amount of randomness or
    “noise” that the market generates. The difference between a real market and a random walk is that the real market has a slight trend component. Perhaps one or two percent of the process is trend, and the rest is noise. That’s the inference problem you’re facing. So the question becomes, how can you use the fact that there is some information in a price series, and how can you extract returns from that information? That’s where testing and optimization come into play.

Don’t Fight the Last War 

  • I’d like to have made more money in the last half of 2008. Whenever we feel that we may have fallen short, we do research to investigate the matter.
  • But it’s important to realize that the last half of last year is simply not enough of a sample to make a substantial change. It would not be a warrantable inference. It would be like fighting the last war.

Strict Testing is Required

  • It has been shown again and again, that without proper controls, even the most honest researcher will unconsciously bias research usually in a favorable direction. Trading systems research is especially rife with possibilities for this kind of wish fulfillment. During more than 20 years, we have seen an amazing variety of ways in which research can mislead or falsify.
  • In response to this we have developed a veritable gauntlet of tests that any system must pass to be taken seriously. We test for post-dictiveness, for computer glitches, and for statistical artifacts. We test for overfitting, for maldistribution of returns, and the degree to which a system takes advantage of unusual and possibly nonrepeatable circumstances. Theses are just a few of the potential sources of trouble that we routinely monitor. This battery of tests can bring runaway enthusiasms back down to earth.

Beware of Summary Statistics

  • Most standard statistical techniques are inappropriate for analyzing trading. Statisticians have developed many delicate techniques that squeeze information from minimal data, but these give false results in this business.
  • I tell traders that if the results don’t sock you in the eye, they’re probably not real. Accordingly, we use only the most robust and assumption free statistical tests. A robust statistical estimator is one that is not perturbed much by mistaken assumptions about the nature of the distribution.
  • We have an aversion to summary statistics that obliterate important structural elements. For assessing systems, we use a technique called bootstrapping so that the complete distribution of past outcomes can make itself felt in decisions; the distribution is not simply viewed in terms of its mean and variance which can give a distorted picture.

Use Yes-No Trade Decision Schemes Rather Than Weighting Schemes

  • It’s a lot easier to look scientific than to be scientific. We try to avoid the kind of delicate fine tuning that gives on the feeling of being very accurate, but that is in fact mostly arbitrary.
  • We have taken to heart the research that shows that simple yes-no schemes, either fully accept or fully reject something, are more useful and more robust than delicate weighting schemes.
  • For instance, we do not favor trades according to how good they are supposed to be, instead we use the following rule: if a trade is good enough to make, it’s good enough to make at full size; if a trade isn’t good enough to make at full size, then don’t make it at all. We adhere to this kind of reasoning all the way down the line. All five systems we currently use are given equal weight. We also try to give equal weight to each of the fifty or so markets we trade.

How to Combine Indicators and Apportion Assets Among Trades

  • The question is: How do you most effectively combine multiple indicators? Based on certain delicate statistical measures, one could assign weights to the various indicators. But this approach tends to be assumption-laden regarding the relationship among the various indicators.
  • In the literature on robust statistics you find that, in most circumstances, the best strategy is not some optimized weighting scheme, but rather weighting each indicator by 1 or 0. In other words, accept or reject. If the indicator is good enough to be used at all, it’s good enough to be weighted equally with the other ones. If it can’t meet that standard, don’t bother with it.
  • The same principle applies to trade selection. How should you apportion your assets among different
    trades? Again, I would argue that the division should be equal. Either a trade is good enough to take, in
    which case it should be implemented at full size, or it’s not worth bothering with at all.

Try to Kill Your System

  • If the performance results of the system don’t sock you in the eye, then it’s probably not worth pursuing. It has to be an outstanding result. Also, if you need delicate, assumption-laden statistical techniques to get superior performance results, then you should be very suspicious of the system’s validity.
  • You have to try to kill your little creation. Try to think of everything that could be wrong with your system, and everything that’s suspicious about it. If you challenge your system by sincerely trying to disprove it, then maybe, just maybe, it’s valid.




No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Copyright © 2005-2014 All Rights Reserved.

Enter your email address to follow this blog and receive notifications of new posts by email.

Blog Stats

  • 397,947 hits
%d bloggers like this: