The previous several posts are what I consider the most critical aspect of sustainable, successful trading. If you didn’t read Risk and Money Management stop here and go back and read them! The principles covered in these posts are the foundation of all of my systems and will be used in system development.
The purpose of this article will be to demonstrate how I develop a system (Note that I spend more time back testing ideas than actual trading). Do NOT expect that I am going to give you a magic system. My personal trading methodology works for me. If I gave readers my current “recipe”, two things would surface - 1) you wouldn’t understand the core why, where, and when because you didn’t put in the time & 2) you would achieve different results than me because my recipe may not fit your personality or trading psychology.
That said, I am willing to guide you to your own recipe.
The Core…
In most asset classes outside of equity options, I am a trend trader. But I am extremely conservative and like to find a reason to “not” trade. I am perfectly happy “not” being in a position, even if I miss the beginning of a trend and hence miss the entire trend.
My core system has five layers, in order of precedence
Trend Shift - using a baseline (single or double)
Momentum Agreement with Trend - using a confirming indicator. This element doubles as an Exit signal if it reverses.
Momentum Agreement with Trend - using a second confirming indicator
Volume Agreement
ALL five layers must agree at the close of a candle (daily, 4 hour, or 1 hour is my personal preference) for me to say GO and place an order.
The first layer was covered in the two previous posts. This post is going to cover the second layer, “Trend Shift”.
Price action for all asset classes (equities, commodities, index futures, forex, crypto, etc…) goes through two primary phases - consolidation and trend. I am most interested in finding a way to understand when a trend is potentially beginning. One of the better ways to define a bullish or bearish trend is via a single line on chart, otherwise known as a moving average or a baseline.
My methodology for Money Management generally gets me through the consolidation phase with small wins and losses, which for the most part cover each other.
For the Trend Shift layer of my system, I am going to make simple [if, then] statement based on the below.
An up trend begins if the price of an asset class closes above the baseline. IF the price on the timeframe of concern closes over the baseline, THEN I will go long.
A down trend begins if the price of an asset class closes below the baseline. IF the price on the timeframe of concern closes over the baseline, THEN I will go short.
Risk and Money Management are per the rules in the Risk & Money Management posts - Position size 2% based on a stop of 1.5 x ATR, take 1/2 profit at 1 x ATR & move stop to open, trail the stop by 1.5 x ATR once the price exceeds open price + 1.5 x ATR.
Exit is based on money management rules or reversing back over the baseline.
With the above rules, I back test a minimum of 100 instances for each type of Trend Shift baseline to determine the win/loss ratio. I do not look at profitability during this phase of system development.
Below are some of my favorite Trend Shift baselines.
Arnaud Legoux Moving Average (ALMA)
The ALMA is based on a moving average indicator and is valid on all time frames. It was designed to address two common drawbacks with the traditional moving averages, responsiveness and smoothness. Anyone who has used a moving average would know that a short term moving average is more responsive, but comes at the risk of being choppy and can result in false signals. On the other hand, a longer term moving average is known to be smoother, but lacks in terms of responsiveness, meaning that price already makes a significant move before the longer term (smoother) moving average catches on.
So traders are generally caught between a fast and responsive but prone to false signals moving average, or bear with the long term smoother moving average which is often delayed when it comes to signals.
The ALMA attempts to show both responsiveness and smoothness at the same time. Interestingly, the ALMA applies the moving average twice, once from left to right and the other from right from left with the process said to eliminate price lag or phase shift significantly, a problem that is common to the traditional moving averages.
Given the fact that the ALMA is more efficient compared to the regular moving averages and the fact that it is nothing but a trend indicator, the simplest way to use the Arnaud Legoux Moving Average is to apply it as a baseline indicator. I like to play with the settings and test for the best win/loss ratio. There is nothing magic in this other than putting in the work. An example of one setting that you can start with…
Here is an example of the ALMA applied to ES (S&P Futures) on the hourly chart starting on Wednesday, December 14 at 9:00 AM Wall Street time.
In this example, there are 4 wins and 2 losses. While the win/loss ratio is poor at 33.33%, there is only one full loss (2% of account position size), three very small losses, and two big wins (2.7% and 2%). Again, I need 100 trades to have a comfort level on the win/loss ratio… That work will be for readers of this post, not only with the above settings but others as well. For ALMA, I generally play around only with the Window Size.
Kaufman’s Adaptive Moving Average (KAMA)
KAMA was developed by American quantitative financial theorist Perry J. Kaufman in 1998. Longer term readers of this publication know that I am big on understanding volatility. Unlike other moving averages, KAMA accounts not only for price action but also for market volatility.
When market volatility is low, KAMA remains near the current market price, but when volatility increases, it will lag behind. What the KAMA indicator aims to do is filter out “market noise.”
Just as the ALMA, Kaufman’s Adaptive Moving Average can also be used to spot the beginning of new trends and pinpoint trend reversal points. I also like to play with the settings and test for the best win/loss ratio. An example of one setting that you can start with for KAMA…
Here is an example of the KAMA applied to ES (S&P Futures) on the hourly chart starting on Wednesday, December 14 at 9:00 AM Wall Street time - the same timeframe as above.
Using KAMA (55) generates 5 trades with two nice sized wins and three very small losses.
Exponential Moving Average (EMA)
This is the simplest and one of the more common types of moving averages. An exponential moving average (EMA) places a greater weight and significance on the most recent data points.
I love Fibonacci numbers, so why not start with a 21 period EMA…
Here is an example of the 21 eMA applied to the same ES (S&P Futures) on the hourly chart starting on Wednesday, December 14 at 9:00 AM Wall Street time.
This is getting interesting… We now have only three trades - two wins and one loss. But again, the win/loss ratio is only valid if I have back tested 100 instances.
Putting it all together…
The use of a baseline in a trend shift is only a partial picture. In all of the above examples, there are instances where the entry rules do not factor consolidation and there are minor losers as price action whip saws back and forth until a new trend is formed. Categorically… I DO NOT rely my entry (or exit) decision only on a baseline. It is only a start.
In the beginning of the post I noted that I prefer a five layer system. This is important enough to repeat again…
Trend Shift - using a baseline (single or double)
Momentum Agreement with Trend - using a confirming indicator. This element doubles as an Exit signal if it reverses.
Momentum Agreement with Trend - using a second confirming indicator
Volume Agreement
ALL five layers must agree at the close of a candle (daily, 4 hour, or 1 hour is my personal preference) for me to say GO and place an order.
For my next post (probably after the New Year) I will begin to cover #3 of my system - Confirming and Exit Indicators. I expect that the Confirming and Exit Indicators will take up multiple posts. And no - I am not going to tell readers exactly which ones I use. But hopefully, you will have enough tools to check them out yourselves.
For premium subscribers, I’ll drill down a bit more and give more focus to my top 5 general system recipes. If you are not a premium subscriber, please consider the following end of year offer…
Keep reading with a 7-day free trial
Subscribe to Green Goose Trader to keep reading this post and get 7 days of free access to the full post archives.