4 Mistakes in Strategy Trading 2
4 Mistakes in Strategy Trading 2
However, in real-world trading the system generated 6 consecutive losing trades and Richard decided to drop it. He found it psychologically difficult to use the method that seemed to have stopped working - even though he knew that the market was being sluggish and the system's behavior was totally natural under the circumstances.
As soon as he stopped using the system, the market entered a growth stage and this trend-following system started working again. As a result an excellent opportunity to earn $35,000 on a single contract was lost!
This costly mistake could have been easily avoided, if they had been trading a portfolio based on uncorrelated markets. It would have ensured a steady profit growth irrespective of the conditions on a single market. Other profits would have nullified 6 losing trades on this particular instrument.
When I asked Richard why he didn't use diversification, he said that the reason was quite simple: the company wasn't paying enough attention to the issue and he didn't have the software to make an efficient portfolio.
The unwillingness to factor in quite a predictable situation as well as the desire to save $2,000 on software resulted in losses 20 times exceeding expenses on the necessary research.
Beat the Markets with Trading Systems ... tinyurl dot com/2vnbsxzu (copy and paste this in a new tab and remove DOT with actual .)
Four Things You Must Do at a Minimum to Protect Yourself from Common Mistakes:
While it's impossible to plan for every problem or emergency, a little proactive analysis and a few simple rules will help you avoid or greatly reduce losses.
Unfortunately, I have found that most beginning and even many experienced traders are NOT conducting any type of analysis, which leaves them completely vulnerable to the types of disasters you just read about. This is primarily for four reasons:
#1. They don't understand the importance of verifying trading systems
#2 They use outdated market analysis techniques instead of adopting the latest and most efficient approaches
#3. Even if they do have a good trading method, they can't use it efficiently
#4. They rely on profits from a single strategy/instrument and don't try to diversify their portfolio
While there are over 20 critical tasks that need to be performed to succeed in trading, I'm going to share with you the 6 that are most important for protecting your capital and creating the most favorable conditions for a profitable and consistent strategy.
Step#1: Don't Trust Any Trading Ideas
I never stop wondering what makes people blindly trust the so-called gurus. I have lots of facts proving that most gurus are nothing more than frauds. However, my main point is that you shouldn't trust anybody, not even yourself. You are the only person responsible for your trading failures and successes. If you fail, the only person you should blame is yourself. If you have a trading idea you must test it. All assumptions and untested ideas cost too much. You simply can't afford it!
Therefore, before you start trading with real money, test your idea under conditions as close to real life as possible. The more accurate the simulation is, the more reliable your tests will be. Be critical and objective when making conclusions. Trust statistics, not your feelings or beliefs.
Step #2: Learn From Those Who Really Know How to Make Money While Trading
Study the practices of those traders who take money out of the markets year after year. Unfortunately, finding such people is a major challenge. Most gurus you'll meet will tell you that they are extremely successful and that they teach trading just for the fun of it or out of pure generosity. I'm afraid that in 99% of these cases all their success stories will be lies.
I only trust statistics coming from unbiased sources. For my analysis I rely on ten-year reports on the best Commodity Trading Advisors (CTAs). Obviously, anybody could be making money for 1-3 years due to sheer luck. Data spanning 1-3 years isn't statistically reliable and can't be trusted. This is why in the ten-year reports, I single out those traders who have been making steady profits for more than 3 years.
Most successful CTAs rely exclusively on mechanical methods and automate their trading to be faster than everybody else. Also, they always diversify their trading.
If you analyze the trends in the algorithm-based trading industry, you will see that most of the solutions out there have been created for institutional traders and cost thousands of dollars. The high demand for algorithm-based trading on the part of institutional traders is only natural. They understand that the best trading opportunities can't last long. Mere seconds separate winners from losers.
oday's markets respond well to arbitrage strategies and high frequency trading. These methods, however, call for reliable mechanical trading systems and high-quality software to deploy them.
I must point out once again that the primitive systems that most newbies get a hold of ceased to work long ago. In the 1970s, even basic trend-following strategies worked fine because the markets weren't as volatile and fast-paced as they are today.
Nowadays, markets require brand-new methods that you won't find in old trading manuals.
Step #3: Run Multiple Tests of Your System under Various Conditions
I am not going to deny the benefits of paper-trading, but I prefer backtesting. Backtesting is the fastest, the most reliable, and most objective way to test a trading method in different situations without letting emotions interfere with your judgment.
When testing your trading idea in real-time on a simulated account, it's impossible to ensure that your tests are error-free and extensive enough to be statistically reliable. I've met few people who would be prepared to paper-trade a strategy for at least 3 months before switching to real-world trading. As a result, their conclusions about the strategy's workability and performance are highly subjective and rash.
Backtesting allows trying a method against different historical data and across different financial instruments. It also ensures that the results are unbiased and consistent. Of course, there are a number of backtesting rules but those must be discussed separately.
Invest just a couple of hundred of dollars to have your strategy coded or spend a few hours of your time to do it yourself and you'll find out if the tested idea is worth anything. I can assure you that you'll reject 99% of the systems that you thought were a sure thing!
Step #4: Don't Miss the Benefits of Optimization for Fear of Curve Fitting
Most people regard optimization with apprehension because in most cases it is applied incorrectly and therefore leads to devastating results. Most people optimize their trading systems to find the best parameters. However, optimization must be approached in a completely different fashion.
How do you create a new trading method? You visually scan data for patterns and check how well those patterns work in various situations. Such eyeball tests are nothing less than implicit optimization.
For example, you will use a moving average with the length of 20 and won't use a moving average with the length of 25. Why? Because you can see that the 25 moving average can't predict market moves as precisely. In other words, you have visually optimized your strategy. The danger of such optimization though is that the chosen value of 20 can be completely random and have no rational foundation.
Optimization is vital for solving two tasks. First of all, eyeballing data for the best parameters is too tedious and time-consuming. Moreover, you might simply never manage to find those best parameters. Let the computer do the job and do it much faster than you would. For example, with the help of the genetic optimization I can test hundreds of indicators with different parameters and find out what works and what doesn't. Visual analysis of the same set of indicators would take centuries.
Second, optimization ensures that the discovered optimal parameters aren't random or over-sensitive to changes. Just create a 3D optimization graph with one click and you'll see how robust your strategy is. If even minor changes affect the strategy's performance and there are no logical explanations for each value, you're simply using an over-optimized
system that might look nice when tested, but will result in disastrous losses in real-world trading.
Step #5: Trade Several Instruments to Ensure Consistent Profits
According to experts, it is vital to trade several uncorrelated instruments. Diversification allows compensating for unfavorable trading periods for a particular instrument while steadily increasing the overall size of your trading account. Obviously, even the most consistent strategy will run into periods of losses. This is the normal dynamics of trading.
At the same time, traders find this phenomenon extremely difficult to deal with. They feel like the system is no longer working because the market has changed. These assumptions can be contrary to reality but more often than not they override logic and common sense and lead to poor decisions.
To eliminate or at least minimize this effect, several strategies must be traded in uncorrelated markets. This way you will ensure a steadier capital growth and abate losses during unfavorable periods. Profits from trading one of the instruments will compensate for the money lost on another.
The result will be a modest but steady growth which is the most important thing in trading. Today, creating an efficient portfolio isn't such a difficult task. Portfolio-level backtesting is now available for a reasonable price. Just a few years ago only large companies with enormous budgets could afford portfolio backtesting.
Step #6: Automate Your Trading Method to Avoid Errors and Routine
It is a well-known fact that a good signal is not enough to enter the market. The latest trading methods call for the best possible entry price. This is especially true for high frequency trading. Human reaction isn't quick enough to respond to price changes within milliseconds. At the same time, the price can change several points which will result in a smaller profit or even in a loss.
Latest trading software products make automated trading perfectly feasible. All you need to do is code your strategy and enable auto-trading. Of course, the trader still must monitor the automated strategy execution and interfere should the situation call for such actions - just as the pilot of an auto-piloted plane may be forced to do.
There is another important reason to automate your trading. There's no point in staring at your computer screen 8 hours a day not to miss a good trading opportunity. And what if you're monitoring a portfolio consisting of dozens or hundreds of instruments?
Instead of wasting your time like that, you could use it to do research and improve your trading methods. Let the computer do all the routine work. This approach will make trading exciting and far less tedious which is particularly important if you want to dedicate most of your time to trading.
Beat the Markets with Trading Systems ... tinyurl dot com/2vnbsxzu (copy and paste this in a new tab and remove DOT with actual .)
