Friday, 27 January 2012

Testing the Delta Phenomenon, Part 2

Towards the end of 2006 the Delta Society message board had a lot of screen shots, short screen casts and uploaded file attachments and quite a lot of discussion about them was generated. As a Delta Phenomenon book owner I had access to this, and I have to admit I was really surprised at how much information was, perhaps inadvertently, being given out. What was typically shown is exemplified by posts #81 and #86 on this forum thread. Unfortunately all that now seems to have been taken down; when I log in to the message board now all I see is two old posts from February 2007 and an "Information for all Clients" section. I remember that back in 2006 I spent hours and hours studying all this information and cross referencing it with the book. The result of all my work was that I now have what are called the Intermediate Term and Medium Term "Delta Solutions" for 30 major US commodity contracts and 6 forex pairs, including the 4 major pairs. However, after getting these solutions, for reasons outlined in my previous post, I did nothing with them and for the last 5 years my hand written notes, along with my copy of the book, have been languishing on my bookshelf.

The main "problem" with the Delta Phenomenon is that it can be very subjective. This assertion is borne out by this quote - "We have observed over time that market participants using Delta and applying it differently do not make buy & sell decisions en-masse at the same time. Also, while the Delta order is perfect, our interpretation of the order is never 100% accurate."  which is taken from the FAQ section of the above mentioned message board. Also here are some random comments that I have cut and pasted from the above linked forum thread (views expressed are those of the original forum posters, not mine)
  • ...as the delta phenomenom could be really nothing but a recurring coincidence that no one can actually use with a good accuracy rate.
  • ...while there may be something in the theory, finding a practical application for it is impossible. To make sense of it everything has to be viewed in hindsight...
  • I thought it was interesting, but just like drawing lines on a chart, I could find a turning point any sequence I decided to use. I think we humans are extraordinary at seeking out patterns in the world around us. Maybe not such a good thing though if there really is no pattern, just perception.
  • Like most any indicator, it looks great in hindsight as you can apply each turning point to the nearest high/low. Problem is, of course, you never know this in real time, only after the fact.
  • Trading with Delta is a lot like looking at an MT4 indicator that "repaints".
  • Mind you, I'm not saying the concept behind Delta is not valid. Just that because of the latitude afforded on either side of the "turning point" day, plus the idea of inversions......it's just real tough to be sure until after the fact. Even the monthly newsletter that Wilder sent out with turning point calls was often subject to correction and change.....a lot of subjectivity with this method.
  • Much is left to the trader's own subjective judgment.
I think readers of this blog will get the point! How can you objectively test something that is so subjective? I don't think merely making or losing money based on subjective interpretation of the points is a fair and objective test of the theory. I posted this forum question and the only answer so far is really good and unless something better is suggested on this forum, or occurs to me, this is the test I plan to implement. A more detailed discussion of this test will be the subject of my next post, but until then readers might be interested in doing some background reading on the matter via the links below
Resampling
Monte Carlo Methods
Bootstrapping
Statistical Hypothesis Testing

No comments: