Market Breadth Primer: CHADTP Past and Present
This article is going to show you an example on the change in characteristics of the Advance / Decline Issues affecting trading models based on the breadth data. The example system I am going to use is called Conner Hayward Advance Decline Trading Pattern (CHADTP). It is originally mentioned in the book Investment Secrets of a Hedge Fund Manager written by Mr. Connor and Mr. Hayward. You can find my review of the book here.
The Performance Over The Past 15 Years
It is much easier to explain what happened to the model by showing you the net gain of the model in a chart.
The top pane is the Emini S&P, 2nd pane is NYSE Advance Issues, 3rd pane is NYSE Decline Issues. All in 390 minute (one full trading session) timeframe.
The bottom pane shows the net gain in dollar amount trading 1 emini S&P contract with the CHADTP system. To simplify the process, only commission of $5 per round turn is taken into account. No slippage is applied.
CHADTP 5 is the CHASTP system that after entry, holding the position for 5 days and exit on close.
CHADTP 6 holds for 6 days.
CHADTP 7 holds for 7 days.
The reason for using 5 to 7 days is that it is mentioned in the book where the best exits happen on the 5th to 7th day after entry.
No stop is used as it is not mentioned in the book thus any such rule added will not be exact implementation of the model given.
The only modification to the system is entry on stop of 0.25 point offset instead of the 0.1 point offset as the original examples in the book is given using the S&P large contract which has its tick size at 0.1 point. Emini S&P tick size is 0.25 point. Thus this change is necessary.
The Damage Done By The Managed Market
As mentioned in my review of the book. It is a book that requires the reader to implement the concepts themselves in order to understand them. This trading setup clearly illustrated the point. The book was published back in year 1995-1996. If you have implemented the model and studied it carefully yourself at the time, you would be trading it, or at least incorporated the ideas into your own strategies. It would make you a lot of money for many years until the end of 2002 to the beginning of 2003.
From that point onward, the behaviour of the breadth data has changed. As mentioned in the article, Market Breadth Primer: Advance / Decline Issues Past and Present, the stock market entered a new era where the number of traded issues in NYSE is no longer unstable. It is as if the market has been managed since. As you can see from the chart, this progressive change of behaviour in advance / decline issues has turned CHADTP, a highly profitable trading model, into something no longer perform. Eventually, the model has broken down and crashed hard during the financial crisis.
Since mid 2011, when the advance / decline issues data entered yet another phase of further tightening in range, something happened. The CHADTP model has come back to life. Unlike its previous performance back in the 1990s where it is stable with consistent gains, this time around the model has more winners and losers of greater magnitude while the mild winners and losers are mostly gone. I am not sure if it is a good thing or bad thing for the model.
But for the time being, it is in a winning streak. It is hot!
Breadth Is Useful Again
My conjecture that a highly controlled market is a predictable market is confirmed by the improvement of this famous trading model based on advance / decline issues.
As I stated in my writings very often, an uncontrolled market will give you a chaotic but predictable market structure. That is the environment we have before mid 2002.
When interventions by the authorities are done whenever they feel like it, which are unpredictable and very disruptive to the auction market process, the damage to the stock market was huge. It turned the market into a mess. That is what we have witnessed from mid 2002 to mid 2011.
Now, as the governments and central banks have cornered themselves into forever money printing, complete order in the market has been restored (in a wicked way) where it is more predictable than ever. Unlike before where the intervention was done randomly in unpredictable manner, the current phase of market intervention is pretty much done everyday and on every significant move. It turns the governments and central bankers from occasional disruptors into active participants of the markets. They are no longer outside forces of the system.
Hence making many market breadth methods useful again.
I have uploaded the CHADTP trading system code to the software store. As usual, I implement simple models in NeoTicker formula for clarity. If you have a version of the system you have implemented for other platform that you are willing to share with other members, you can post that in the download area.