PowerZone Trading

Indicators versus strategies

What is the difference between an indicator and a strategy?

 

Indicators, such as MACD and stochastic, are used in technical analysis to predict changes in price trends and patterns. The goal of using indicators is to define high probability trading setups. Indicators are mathematical calculations used to highlight what has already happened and what is currently happening to allow us to make an educated guess about what will happen next. For instance, a moving average crossover often predicts a change in price trend. In short, indicators help us determine areas of price that could be profitable entry and exit points in our trades.

 

Strategies may also utilize the calculations of indicators to create a defined set of rules for trade entries and exits. Strategies, however, outline exactly how we are going to use the indicators. For instance, a moving average crossover can predict a change in price trend. But how exactly do we quantify that? A strategy would state that: the system should use a market order to buy 500 shares on the first price bar where the 50-period simple moving average (SMA) crosses above the 200-period SMA. The indicator finds the ideal setups; the strategy plucks information from the indicator(s), establishes a position size and specifies an order type, then pulls the trigger. 

 

When you use a publicly available indicator or purchase a proprietary indicator, you have a tool that will help you make trading decisions. The indicators themselves do not create entry or exit signals. You will need to do your homework (research, backtest, paper trade, etc) to determine how to use the indicator(s) effectively in your own trading.

 

When you use a strategy, a set of rules defines how any indicators are used to define trade entries and exits. Because strategies are necessarily objective, backtesting can show the expectancy results.

Working with a Programmer

 

Most traders, whether greenhorns or an experienced fund managers, have ideas for trading indicators and/or strategies that they would like to test to determine how they would perform in the markets. Going through years of data and locating potential trading opportunities by hand is time consuming and, due to its tediousness and subjectivity, can yield flawed results. Developing these ideas into code can let our computers do the hard work: backtesting, multi-variable optimizations, strategy performance reports, etc.

 

Not all traders are capable programmers, however. Nor do all aspire to be. This is where an experienced programmer can help turn a trading idea into a testable, quantifiable trading tool.

 

Most trading platforms have a proprietary language that must be studied and learned over the course of years to develop elegance in coding that language. Finding a programmer experienced in the particular language is the first step. Many trading platforms' websites have a list of specialists who can provide custom programming.

 

  

TradeStation has a list of EasyLanguage Specialists who have demonstrated to TradeStation that they are experienced, professional developers with a deep understanding of TS EasyLanguage.

 

Your initial contact with the developer can be a quick email stating the basics of your project ("I am looking to create a simple strategy based on a moving average crossover...") to find out if the developer has the time and expertise to work on your particular project.

 

If the developer indicates that he or she can accept your project, you can request a Non-disclosure Agreement if you want to protect your trading idea before releasing more details. After you send a detailed project outline (please read Tips below), the programmer will return a time and cost estimate for the project. If this is acceptable, you can give them the go ahead for the project. Some developers will require some type of down payment (usually 50% of the estimate) upfront. Others will bill after the project is complete.

 

Plan on reviewing and testing the indicator/strategy when the developer sends the first draft. Unless the project was very basic and straightforward, some type of revision is to be expected. Either your specifications were not 100% clear (even though they were clear to you) or the developer interpreted your idea differently than you expected. Most developers do not want unhappy customers and will work with you to provide a solution with the various tweaks and revisions. You can also request a consultation session to go over the variables and nuances of the indicator/strategy as programmed. This will typically be billed separately from the actual programming.

 

Once the project meets your specifications, you can thank and pay your developer, and start enjoying your shiny new code.

 

Tips:

Any ambiguity in your project specs will only cause frustration (to you and your programmer) and ultimately more money in development costs. Almost any idea that is quantifiable can be programmed (depending on the limitations of the platform). You will do yourself a favor if you provide only absolute rules to the developer. Here's an example:

 

 
An example of concise entry and exit logic.

 

This is just an example...it shows definite entry and exit logic. It is also helpful to include any specific markets and intervals that you plan on using (for example, the e-mini Russell, 5 minute chart). If you have pages and pages of rules, try to narrow it down to one page for the initial project. You can then add filters through additional programming. This tends to be more cost effective (and easier to troubleshoot) than having a programmer sort through pages of ifs and thens.

 

Working with a seasoned developer can be a great experience. You get to turn your ideas into functional tools without having to learn your platform's proprietary programming language. Be as clear and concise as possible, be completely objective in defining your project, and plan on some time for revisions to get your indicator or strategy performing as you expected.

Trailing Stops

Trailing stops are dynamic stop orders that follow price in an attempt to lock in profits, while allowing a position to capture additional profits during a price trend. A trailing stop will move up in a long trade to follow the rising market and will move down in a short trade to follow a falling market. The trailing stop can be based on a certain dollar amount: for example, a trailing stop can follow price and allow the trade to fall back $100 from its most profitable point before it is closed out. A trailing stop can also be based on percentage: the stop level adjusts to a specified percentage below the market price (for long trades) or above the market (for short trades).

 

This chart of McGraw Hill (MHP) from September 3, 2009 shows a downtrend with a trailing stop applied. I used a volatility based stop to set the levels for my trailing stops. This was a beautiful downtrend that allowed the trailing stops to work very effectively. The trailing stop levels are shown here as small red dots above the price bars. The blue dots signify where the trailing stop is touched, and therefore, the trade closed.

 

McGraw Hill on September 3, 2009. A volatility based stop was used to lock in profits along the nice downtrend. I have a paintbar study applied to the price bars indicating that volatility shows downwards pressure.

 

Next, let's look at an example where the trend is not as enduring and see how trailing stop levels can still be used successfully.

 

 E-mini Russell on September 15, 2009. A trailing stop was used to lock in profits on both upward and downward moves. I have a paintbar study applied to the price bars indicating where volatility shows upward momentum(green bars) or downward  pressure (red bars).

  

In this chart example, we don't have the long-lasting trend as seen in the MHP chart. Instead we have a non-trending market where we are still able to effectively use trailing stops. The green dots show my trailing stop for a long position; the red dots indicate the trailing stop level during a short position.

 

I find a useful method of using trailing stops is to close out the first part of a trade with a profit target and allowing the remaining shares/contracts to exit using the trailing stop levels. In this manner, you can allow the trade significant movement during trending markets while not risking the entire position.

Forward Performance Testing

Forward performance testing, commonly referred to as paper trading, is an important step in evaluating a trading system. Once we have determined that the performance metrics look reasonable in our backtesting with both in- and out-of-sample data, we are ready for forward performance testing.

 

Forward performance testing is the simulation of actual trading. Trades are taken on paper only, so any profit or loss is hypothetical. The trading plan should be followed exactly, including how the orders are placed. Many trading platforms, such as NinjaTrader and TradeStation, allow users to trade simulated accounts. Traders are able to watch live market data and place realistic trades, without risking any money.

 

One of the biggest challenges here is simulating a real trading atmosphere. Obviously it is much easier to play than to commit: trading for pretend and trading for real are two different animals. But we want to try to create the same types of pressure we feel when trading with real money. Here are some ways to keep your paper trading realistic:

 

· Document all trades

· Do not reset trades if you are using a trading simulator (such as in TradeStation)

· Do not rationalize a losing trade by saying "I would have never taken that trade in real life"

· Pay close attention to order entry (you can learn a lot here)

· Take good notes. If something happens (you went long instead of short) write it down and learn from the mistakes

· Follow your trading plan. How else are you going to tell if the system is profitable?

 

A note on order entry:

I cannot stress how important it is to practice order entry. It is easy to get confused and place the opposite order than you intended. And it is easy to get sidetracked by your winning trade and forget to place a protective stop order in the market. Check your open orders often and be certain that you still want those orders in the market (otherwise cancel the order). The more you practice placing orders, the easier it will be to place actual trades in a live market without making costly mistakes.

 

Once you are comfortable placing trades in the simulated market, and your trading plan is performing as expected, it may be time to consider going live. Start small (one contract, for example) and follow your plan exactly. Remember, if you deviate from your plan, you all but eliminate any expectancy that the system provided.

More on Optimization

In my last blog entry, I outlined several key performance metrics in defining the efficacy of a trading system: maximum drawdown, total net profit, profit factor, percent profitable, and average trade net profit. While it is easy to become focused on the outstanding performance of one of these metrics, it is critical to evaluate the metrics as a whole.

 

As an example, say we design a system that shows an average trade net profit of $200. At first glance this looks like a winning system. But what if we also look at the maximum drawdown and learn that the system would suffer incredible losses (bigger than our account size or risk tolerance). As another example, imagine we create a system that is 80% profitable. That sounds good, right? But what if the average trade net profit is a negative number? It can happen if the average loss is substantially higher than the average win. It's important to consider how each of the metrics is connected to the others.

 

When we optimize our systems we intend to improve the performance of the system. We need to be careful, however, not to over-optimize our systems. Over-optimizing refers to excessive curve fitting of data producing a trading plan that will be unreliable in actual trading. An over-optimized system will often create excellent results on historical data (through back-testing) but will more than likely perform poorly during any out of sample testing or live trading.

 

Fortunately, there are things that we can look for to help us determine if we have over-optimized a system:

 

    - Performance metrics that are too good to be true (like profit factors in the double digits or percent profitable reading above 80%)

    - Performance metrics that are based on very few trades

    - The absence of correlation between in sample and out of sample trading periods

 

Out of Sample vs. In Sample data

 

One way to help evaluate a system is to set aside "out of sample" data on which to test the system after any optimizations have been completed. No optimizing should occur on the out of sample data. Ideally we will have lots of trade data to test and study when designing a system. If we can do our optimizing on about 2/3 of the data available, reserving the last third as out of sample, we can confirm the results of our optimized system on this trade data. If the system performs well during this time period (on the out of sample data) that is a good sign that the system may be viable.

 

Next week we will look at forward performance testing, the next step in confirming the efficacy of a system.

Finding Balance in Strategies

Finding Balance in Strategies

One challenge system developers face is finding the magic set of rules that will perform reasonably well in a range of market conditions. It can be easy to fixate on optimistic report statistics while losing track of trends in the overall market that may have contributed to the numbers. As the markets are fluid and dynamic, with uptrends and downtrends, our trading systems should be able to handle all that the markets can throw at them.

The overall downtrend in the markets over the past couple of years can create a short-side bias for trading systems - especially trend-following systems. Imagine if your system only went short - you would have probably done pretty well! However, chances are your system would not have done very well if it had been designed to just go long. It is important to find a balance that allows a system to perform well under both bear and bull market phases.

When we design systems, we have the advantage of being able to apply logic to the past x number of years and see how the system would have done: wow, hindsight looks great!

But, even with the past data, it is important to weight the immediate and futures markets as well, even if it's just a guess. We can't assume the markets will continue to go down, just like we can't assume the markets would go up forever. We can assume the markets will move, so having a balanced system deserves some thought.

Take a simple stop-and-reverse 17-day breakout system: if the current day closes higher than the previous 17 days, the system goes long; if the current day closes lower than the previous 17 days, the system reverses and goes short. Figure 1 shows the trades this system took with a definite bias towards the short positions. Look at the performance report in Figure 2. Had we tried this as a long-only system we would have closed up shop. But the shorts kept us afloat. But can we sit back and relax? Probably not. Our short trades are winning much more frequently than our long trades, so I would be worried about how the system would be affected by a change in the market trend.

This is just a thought...

Consider trying to achieve a balance in your trend-following systems so that they can reap the rewards in both bear and bull markets.

Understanding Leverage to Maximize Returns

The following excerpt was taken our book, Make Money Trading: How to Build a Winning Trading Business. The section that follows is a short discussion on understanding the level of leverage that maximizes returns.

 

Lee Leibfarth is an independent futures trader who designs, tests and implements his own trading systems. He is an affiliate of the Market Technicians Association, and founder of PowerZone Trading. His articles on trading have been featured in The Technical Analysis of Stocks & Commodities, Futures, Active Trader, and Trader’s Journal magazines.

 

Jean Folger is a stock and futures trader who began her trading business while searching for dynamic portfolio diversification. She discovered day trading and has become highly adept at analyzing and trading the intraday markets. Her articles have been featured in Futures magazine. Jean specializes in teaching and mentoring for women. Their book is available here.

Excerpt from Chapter 7: Developing a Trading Plan

Leverage and Margin

Deciding precisely how much to trade is a key aspect of a trading business that allows traders to control risk and maximize profits. The amount of margin, or leverage, that a trader chooses to use determines how much he or she can buy or sell at any given time. Traders must decide how to use this powerful tool most effectively without creating a high probability of ruin for their trading businesses.

It is recommended that traders begin trading only with a small percentage of the amount of money that is available in their accounts. Traders should start small and increase their position sizes as their profits and trading skills grow. Newer traders that are too anxious to trade larger positions (using the entire value of their accounts and/or large amounts of leverage) almost always fail. Margin and leveraging should be applied with extreme caution.

A trader can determine, for example, the maximum number of shares of QQQQs (trading at $44.20 per share) they could afford to buy with a $60,000 trading account, by dividing the account size by the current value of the stock.

Max Position Size (Stocks) = Account Size / Stock Value
                        = $60,000 / $44.20 = 1,357 Shares

A standard overnight margin account allows 2:1 leverage for stocks, and would give this trader the buying power to purchase twice as many shares than the account size could otherwise afford. This brings the maximum marginable position size up to 2,714 shares (Max Position Size * Margin Rate). It is advised that traders should never trade at maximum margin levels.

A more responsible method to determine a position size is to trade a percentage of the overall marginable value of an account, instead of the complete account value. For example, a trader may choose to base the position size on 30% of the margin account value:

Max% Position Size (Stocks) = (Account Size * Margin Rate * Trading Percentage) / Stock Value
                        = ($60,000 * 2 * 0.3) / $44.20 = 814 Shares

This amount is typically rounded down to the nearest 100 shares to allow for easier order entry and trade management. Accordingly, the above example would relate to 800 shares of the QQQQs.

It should be noted that the futures and forex markets require a necessary amount of margin and this should be factored into any type of position sizing. For instance, trading 30% of a maximum position in the E-mini Russell 2000 futures contract (initial margin of $3,500 per contract) with a $60,000 futures account could be calculated as follows:

Max% Position Size (Futures) = (Account Size * Trading Percentage) / Initial Contract Margin
                        = ($60,000 * 0.3) / $3,500 = 5 contracts

As stated above, deciding how to establish a position size is a critical factor in building a trading plan. Two methods are commonly used by traders: constant position sizing and fixed percentage position sizing. Constant position sizing uses the same amount of shares, contracts or currency in every trade, while fixed percentage position sizing varies the position size as the value of the account changes.

A comparison of these two position sizing methods can be seen in Figure 7.2. The equity curve for each of these position sizing models is displayed using a sample trading plan applied to 6 years of historical data from the E-mini Russell 2000 futures contract. An initial account balance of $60,000 was used to construct these equity curves. While the constant position size model would have traded with a consistent five contracts, the fixed percentage model began with five contracts and increased the position size as the account grew. This progressive method of fixed percentage position sizing is one of the key concepts contributing to the effectiveness of the Turtles’ trading plan. This essentially allows traders to compound their trading profits and risk more only after they have made more. The previous equations for max% position size can be used to calculate a position size based on the current account value, and should be calculated following every trade.

 

While fixed percentage position sizing clearly has the advantage, constant position sizing is still useful to traders. This method can work well for traders who are just getting started, or trading smaller accounts. Constant position sizing should also be used during the testing and evaluation of a trading plan.