Turtle Trading Rules

Chapter 26 Defense System (1)
Trading is not a race, it is a boxing.The market will hit you with an iron fist and do everything in its power to beat you.To win, you have to be in the ring safe and sound when the bell rings for the end of 12 rounds.

When novices design trading systems, they always want to rely on historical tests to find an invincible super system.They believe that systems that have performed well in past tests will also do well in the future.If they found in the test that the CAGR and MAR of a certain system (temporarily called the Omega system) were 10% and 0.2% higher than another system (temporarily called the Alpha system), they would definitely choose Omega.In their opinion, since Omega looks so much better than Alpha, it would be foolish to choose Alpha.

With experience, you will realize that there is no perfect system in this world.Omega may perform better in certain market conditions, and since the market conditions most favorable to Omega have prevailed for some time in the past, it may perform far better than the alpha system in historical tests.But unfortunately, no one can guarantee that such a market state will occupy the mainstream in the future.In other words, future market conditions may be different from those of the past.Therefore, if the difference between Omega and Alpha in the test is only determined by a certain distribution law of the market state, then once the market law changes in the future, this difference may no longer exist.

for example.Assume that Omega outperforms Alpha in calmer trends, and Alpha outperforms Omega in volatile trends; of the 20-year test period, the market was dominated by calmer trends in 13 years and only 7 Years are dominated by fluctuating trends. If the same distribution pattern reappears in the future, Omega will perform better than Alpha in the future.

But what if five of those seven-year volatility trends occurred in the past 7 years?What if the trader effect brings about a change in market behavior such that future trends become more volatile?If so, the future winner is more likely to be Alpha, as it performs better in volatile trends.Conversely, what if markets were cyclical, alternating between periods of calm and volatility?Doesn't this mean that Omega will be even better in the future?Because when the market returns to calm again from its recent volatility, Omega outperforms Alpha.

unpredictable future

In many cases, we will not be able to make confident decisions because we do not have enough information.The reason is the lack of data.Consider the serial symbol "QQQVVQ": if this represents the alternation of quiet periods (Q) and volatile periods (V), can we determine from a distance that the future market is more likely to be in a calm period or a volatile period?If you've read the previous chapters carefully, you'll realize that we haven't been able to draw any firm conclusions from such a small sample.Even if we saw a larger sequence, like "VQQVQVVQQQQVVQ," and that looked like a periodic shift, we still wouldn't have enough data to draw that conclusion.

In cases like these, it's best to accept that we don't have enough data to tell what the future holds.Therefore, we cannot accurately predict the relative performance of the system in the future, and can only speak in general terms at best.A mature understanding of this is essential to building a robust trading system, and like many other issues in the trading world, recognizing reality is a critical first step.Once you recognize reality, you can make realistic decisions and adjust your behavior accordingly.

Two characteristics of a sound trading strategy

The so-called stable trading is to use a stable trading strategy to resist the risk of market fluctuations.To do this, you must first accept that no one can predict the future, and that any test based on historical data is inherently biased.

Ironically, once your trading strategy takes future unknowability into account, your performance becomes more predictable.This may seem contradictory, but the reason is simple: If your trading strategy is based on the assumption that the future is unknown, then any future market state is already in your expectation, and you don't need to predict anything.Conversely, if your trading strategy is based on certain assumptions about market characteristics (any assumption is true), then if those assumptions fail to hold, your strategy will lose its footing.

So, how do we develop a trading strategy that does not depend on specific market conditions?There are two main characteristics of any sound trading strategy: diversification and simplification.When it comes to the contribution of these two elements to robustness, our nature provides the best example.In this regard, we can vividly compare the robustness of the trading system to the viability of the ecosystem and each species within the system.

Decentralization

In an ecosystem, nature does not rely on just one or two species to complete a task.It wouldn't just have one carnivore, one food source, one herbivore, one scavenger that would clean up the carcass.Decentralization is very important because it prevents the ecosystem from becoming passive in the event of a drastic change in the number of certain organisms.

Hatching

In stable environments, complex ecosystems are more resilient, and complex species appear to far outperform simple ones.But during times of change, complex species are more likely to perish, because at these times the hardiest species are those that are very simple, such as viruses and bacteria.The reason why simple life forms are stronger is that they are not so dependent on special environments.Whether it's a meteor hitting the Earth or a volcanic eruption causing a dramatic drop in temperature, simplicity can be a huge advantage when these kinds of upheavals are taking place in an ecosystem.Conversely, if the climate changed, dependence on the old climate would be a huge disadvantage.

There are also complex species that are very hardy and can survive a wide variety of conditions.These species generally evolved in variable climates or conditions and thus developed the ability to adapt to change.These strong species are our role models for building robust trading systems.

Now that we have considered two elements of robustness - diversification and simplification, let's see how we can strengthen our trading strategy from these two aspects.Minimizing laws that exacerbate dependence on special market conditions is the key to simplification.To increase diversification, we can choose as many uncorrelated markets as possible, or use several different trading systems at the same time-this way, no matter what the future market conditions are, there will always be some in your portfolio that will perform well system.

Strengthen the robustness of the system

There are two main points to strengthen the robustness of the system. One is to ensure that the system rules can adapt to various market conditions, and the other is to keep the system simple and not easily affected by market changes.

You can increase the robustness of your system by enhancing its ability to adapt to different market conditions.Such systems are super adaptable, just like those organisms in nature that have the ability to survive in variable environments.Humans are an example.People can live in the Sahara Desert or in the ice and snow of the Arctic, because human beings' high intelligence enables them to adapt to these extremely different environments.

Any system has a market state that is more suitable for itself.Trend-following systems perform better in calm trends, and counter-trend systems perform better in stable and volatile market conditions.The reason why a filter can make a system more robust is because it can eliminate markets that are in an unfavorable state.The Donchian Trend System's Trend Combination filter is an example.This filter prohibits trading on breakouts that are in the opposite direction of the general trend, which can only happen in unfavorable market conditions.Breakthroughs in the direction of the general trend are more likely to occur in a trending market, and adding this filter will make the system more robust.

Likewise, simple laws increase the robustness of a system precisely because such laws work in a wide variety of different situations.Complex systems generally become so complex because system developers are always trying to add new laws that take advantage of certain market conditions or behavior that they notice.The more such laws there are, the more dependent the system is on specific market conditions or behavior.This makes it more likely that future markets will not have these conditions, and in such a market, these complex laws will no longer be in effect.

Compared with these complex rules that are "tailor-made" according to special market behavior, simple rules based on more basic concepts are more applicable in actual trading.In the long run, simpler systems are more viable.

Choose from multiple different markets

Selecting multiple different markets is one of the most effective ways to increase the robustness of your trades.The more markets there are, the more likely you are to encounter favorable conditions in at least one of them.For trend following systems, the more markets you participate in, the more likely there is a trend in one of them.

This means that you should include as many markets as possible in your portfolio, and there must be new opportunities emerging in these markets, so they cannot be too correlated with each other.For example, there are several U.S. short-term interest rate products that almost always rise and fall in tandem, and choosing multiple such markets does not improve diversification.

If your system does not require close monitoring of market movements from morning to night, you may consider offshore markets.These markets can go a long way toward decentralization, increasing the robustness and consistency of transactions.For some of the systems in this book that buy at the open based on closing price data, the global markets are a much better place to use it, because if all you care about is the closing and opening prices, the jet lag is a problem. Not so important anymore.

How to choose a market
TradeStation automated trading software is by far the most popular system testing platform, but it has a huge limitation: it cannot test multiple markets at the same time.This has a side effect: Many traders look at individual markets rather than entire portfolios.Because of this, many people mistakenly believe that certain markets cannot be included in a portfolio on the grounds that they are unprofitable or underperform other markets.

There are two problems with this view.One, some markets may only trend every few years, so short-term tests of 5-10 years may not adequately reflect their potential.Second, although some markets have unsatisfactory profit levels, they have great diversification value, and the advantages of the two outweigh the disadvantages.

(End of this chapter)