Have you ever wondered about the historical accuracy of the Bradley Siderograph, including its ability to identify important turning points in the market? Also, have you ever wondered why the Bradley Siderograph works better in some years than in others? I have wondered these questions myself, so I set out to answer them. What I found blew me away, especially with respect to the ability of the Middle Terms to identify turning points in the market. See below for details.
Backtesting the Bradley Siderograph
I analyzed the Bradley Siderograph Turn Dates with a “Bradley Bars” Power of 100 out of 100 against the Dow Jones Industrial Average (DJIA) from 1900-2014. Click here for an overview of the Bradley Bars and how they work. The preliminary results of the backtesting indicate that Bradley Turn Dates are associated with a greater likelihood of a change in the market’s trend.
There were 158 Bradley Turn Dates with a Power of 100/100 during this time period. For each of these turn dates I calculated the ratio of the performance of the Dow Jones Industrial Average 1 day before and after the Turn Date, 2 days before and after the Turn Date, 3 days before and after the Turn Date, etc. all the way up to 30 days before and after the turn date.
I divided the performance after the Turn Date by the performance before the Turn Date. A positive value would be indicative of no change in the trend (i.e., a continuation of the previous trend), and a negative value would be indicative of a change in trend. For example, if the DJIA was up 2% for the 10 days prior to the turn date and then it was down 1% in the 10 days subsequent to the turn date, the value would be -50% (i.e., [(-1%)/([(2%) = -50%]). If the Bradley Siderograph is effective at identifying turn dates, we would expect the result of this formula to show an overall negative median value.
For all 41,762 days calculated for the Dow Jones Industrial Average, the overall median value using the formula above was 7.11% for 15 days before and after a particular date. This means that on average the Dow Jones Industrial Average tends to continue its overall trend.
However, for the 158 Bradley Turn Dates with a Bradley Bars Power of 100/100 between 1900-2014, on average the median was -17%. In other words, for the period 1900-2014, Bradley Siderograph Turn Dates with a Bradley Bars Power of 100/100 was more likely to result in a change in trend than the average day. Not only was the Bradley Siderograph’s value negative as we would have expected if it is working, it was also more than twice the value of the average trend continuation for the DJIA.
Middle Terms Backtesting
Although the results above are favorable with respect to the effectiveness of the overall Bradley Siderograph itself, the results for the Middle Terms are even more impressive! The Middle Terms are one of the three elements that, when added together, equal the Bradley Siderograph. If you would like to learn more about how the Middle Terms relate to the Bradley Siderograph, you can click here to learn more.
Between 1900-2014 there were six Middle Terms Turn Dates with a 100/100 Bradley Bars Power. All six of these dates resulted in a change in trend.
Why are the Bradley Turn Dates Stronger or Weaker in a Particular Year?
The Bradley Turn Dates performed very well during 2014, but how well will they work 2015? If you have followed the this indicator over the years, you may have noticed that some years it appears to perform well, whereas its performance is weaker in others. One factor that could result in “stronger” or “weaker” years for this indicator is the Bradley Bars Power of the turn dates. For example, in 2014 there were three Turn Dates with a power of 100 out of 100. Also, during 2014 there were especially powerful turn dates in the Middle Terms, Long Terms, and Declinations (i.e., the three elements that combine to create the Bradley Siderograph). However, some years have turn dates with a lower Power, which results in Turn Dates that aren’t as strong.