General Principles for Beginning or Struggling Traders

Although increased leverage in the forex market, in relation to other markets, allows new traders having less capital to get their feet wet, it can also result in drowning. One of the most important features of automated back testing is its automated drawdown calculations. A nice profit factor can be utterly destroyed by a high drawdown coupled with an undercapitalized account. In the forex market, the crosses (minor pairs) that have the highest volatility will carry lower leverage. Again, think GBPJPY. I believe that this is a good thing in terms of learning to trade responsibly.

Broker-Dealer Business Models

On a related note, I have noticed that many forex broker-dealers that offer sky-high leverage generally are not connected to the prime (interbank) market. Instead, they tend to “clear” retail traders’ orders in-house and/or by way of their partnered market makers/liquidity providers (hedge funds, investment banks, etc.). While it is true that there is no centralized spot forex market, it is also true that prices can vary substantially between the greater prime market and these smaller captive markets.

Forex broker-dealers that use the captive market model sometimes offer up to 1:2000 leverage, while a broker-dealer that uses the interbank market model generally offers up to 1:50 leverage. One potential explanation that I have found is that the market makers/liquidity providers pay the captive model broker-dealers for each trade executed, so those broker-dealers have a vested interest in encouraging their retail customers to trade as much as possible.

Regarding other markets:

• Contracts-for-difference (CFD’s), by definition, are not connected to any underlying market (captive model);
• Stock broker-dealers that are connected to centralized exchanges may use the captive model as well;
• Options broker-dealers that are connected to centralized exchanges may use the captive model as well;
• Electronically traded fund (ETF) broker-dealers that are connected to centralized exchanges may use the captive model as well;
• Futures broker-dealers that are connected to centralized exchanges are the least likely to use the captive model; and
• Forex broker-dealers that are connected to the interbank market may use the captive model as well.

To figure out which business model(s) that your broker-dealer employs, go through your trading agreement with a fine-toothed comb. Also, carefully review all of your broker-dealer’s published disclosures. Many jurisdictions require certain documents to be provided to customers and/or published online. If this is the case in your jurisdiction, a list of your broker-dealers’ business partners will almost certainly appear in one or more of such documents.

Ultimately, we as retail traders must weigh the vested interests of a broker-dealer against our need for the best execution of smaller retail orders. Smaller orders receive worse execution if they are sent directly to the interbank market, where the usual institutional trades are much larger. Perhaps nothing is more important here than the integrity of the people who operate your broker-dealer. If you can manage to sort out honest online reviews from seeded online reviews, I leave that to you.

Coding

While the ability to code/program is not required to trade, it has several inherent advantages:

• It is much easier and efficient to back test algorithmic strategies. Although manual back testing is possible as described above, manual back testing becomes too burdensome to generate the plethora of statistics that an automated back tester generates. Now, let’s say that you complete a manual back test, and then you change an input setting. Then you have to start the entire manual back test over again. This brings us to the issue of optimization. An automated back tester can generally run the strategy through a multitude of markets, instruments, and input settings relatively quickly. In contrast, this greater process is unworkable manually.
• Futures and forex markets are basically open 24 hours per day and 5 days per week. This does not imply that anyone should be trading all 24 hours. I generally avoid trading at least the several hours surrounding the daily market open/close. Unless you’re gaming the gaps or wealthy enough to game swaps (carry), the bid-ask spreads are too wide and liquidity is too low to trade at that time. However, there are often strong price moves (volatility) during the Tokyo, London, and New York sessions… but not in all markets and instruments. GBPJPY is good for those sessions excluding 3 or 4 hours between liquid New York hours and liquid Tokyo hours. In contrast, manual traders obviously can’t trade all of those hours without sleeping enough hours daily. Furthermore, attempting to do so will tempt the manual trader to use more margin to increase profits in limited trade opportunities–while the auto-trader will maintain a smaller trade size to exploit more trade opportunities.
• Feelings/emotions are 100% excluded from pure auto-trading which is highly beneficial in light of the aforementioned drawbacks of intuitive trading.
• Another experienced trader once told me that if I have a statistically successful strategy, my goal should be to follow the rules of my strategy instead of focusing on the profit/loss of any particular trade. The point here is that a successful strategy can lose every 4 out of 5 trades if the average winner is for example, 5 or 6 times the absolute value of the average loss. This is much easier for a robot to endure than it is for a human to endure (see the aforementioned intuitive/feelings/emotions discussions).
• Proficient coding ability allows you to implement almost any and every strategy that comes to mind fairly efficiently, which helps you rapidly determine whether the strategy shows promise or not (see the aforementioned back testing and optimization discussions).
• Coding can provide its own sense of accomplishment to a coder, even if the code written merely trades as the coder intended. You could say that even if the strategy logic were unprofitable, the code operated correctly. After becoming a proficient coder, successful trading can morph into somewhat of a side effect of programming. A break-even algorithm can potentially be code-tweaked and optimized into a winning strategy.

Spaghetti Charts

In my experience, there is no need to use 100 different indicators to design a successful trading strategy. I have created successful strategies that use only 2 or 3 indicators, including a strategy simply based on 2 moving averages and 1 oscillator. There is a lot more going on within a single indicator than some traders might realize. For example, a simple moving average (SMA) can be analyzed based on:

• Its position in relation to price,
• Its slope (up or down),
• Its angle (in degrees),
• A bar shifted (displaced) value, and/or
• A series of its values across multiple bars that forms a pattern.

If we add in another SMA indicator, then we have twice as many attributes to potentially analyze… plus potential evaluation of the SMA’s in relation to each other. If we throw in the moving average convergence-divergence oscillator (MACD), now we have 2 more line studies, a histogram, and a zero level (and all of their respective attributes like above) for potential analysis. One must begin to wonder how many indicators a trader really needs, and to what extent the human eye/brain can to compete with a robot in tracking all of these strategic-variables on such a microscopic scale.

Arguably, Renko bricks are not only a chart type but also an indicator and I have successfully used Renko bricks as a standalone utility. Of course, coders can combine a multitude of indicators into one indicator that cleanly and unmistakably provides entry and exit signals which I have also done. In contrast, if I had attached all of those indicators to a chart as a manual trader, I would not have been able to see through the spaghetti.