You will notice a new layout for the Great Trough Detector (GTR) chart as shown below. The Signal Panel has been revised to be more intuitive and we have added a new count panel below that:
1. The Signal Panel
The first task of this panel is to highlight your two main opportunities:
- Class-B : When the 13Wk New Hi/Lo breadth signal prints a reading below 27.4 “Correction Warning”
- Class-A : When the 13Wk New Hi/Lo breadth signal prints a reading below 10.0 “Deep Trough”
Class-B opportunities will always precede Class-A opportunities, and a Class-A opportunity always cancels out a current Class-B opportunity. Not every Class-B opportunity becomes a Class-A opportunity.
The duration of the respective opportunities is defined as follows, after which they expire:
- 5 days for the Class-B’s
- 10 days for the Class-A’s
The second taks of this panel is to alert you of BUY signals. If during the course of a respective opportunity duration, the Hi/Lo breadth signal punches above 80 and has spent at least 4 days below 80 prior to that, we generate the respective Class-A or Class-B BUY-signal shown by green or orange “spikes” on the signal panel. Once a BUY signal has been generated, the associated opportunity signal is cancelled.
2. The Counting Panel
This appears at the bottom as a new panel and is the result of further research we have concluded on the nature of the 13Wk New Hi/Lo breadth signal. Whilst 90% of all visitations of the breadth signal below 10 eventually lead to a Class-A BUY signal, we can further complement the resulting Class-A buy signal with a rolling count of the number of days the signal spent below 10. Class-A signals accompanied by higher “days below 10” counts are associated with larger, more intense and protracted corrections and offer higher confidence/rewards. The 27 year historic counts of all 67 visitations of the breadth signal below 10 are displayed below:
Single-day visitations below 10 occurred 22 of the 67 times or 34.4% of the time. However there were only 5 visitations below 10 that lasted exactly 2 days meaning a total of 22+5=27 occurrences (42.2% of the sample) spent 1 or 2 days below 10. Similarly, some 57.8% of the entire historical sample (22+5+10 occurrences) spent no more than 3 days below 10. So counts of 3 or more represent more than 50% of the sample set and counts of six or more represent less than 25% of the sample set.
There are horizontal dotted lines in the count panel at the +3 and +6 levels to highlight these groupings and we can thus categorize our Great Trough Signals further as below:
- SMALL : 1-3 days spent below 10 (57.8% of the signals)
- MEDIUM : 4-6 days spent below 10 (20.3% of the signals)
- LARGE : more than 6 days spent below zero (21.9% of the signals)
As of 8th September 2014, the count was 8 which means the current correction it represents is LARGE. According to the distribution table, counts of more than this represent only (100-84.4)=15.6% of the entire historical sample set. Conversely we can make a rough conclusion that there is a 84.4% probability we have seen the last visitation below 10 in this current correction, assuming historical trends persist. Counts of ten or more fall among only 10% of the sample set (clearly outliers) and thus should offer 90% plus probabilities you have witnessed the last visitation of the breadth signal below 10.
About Intra-day Signalling
The Great Trough chart is updated every 15 minutes during the course of a trading day. For this reason you may see the signal panel and the count panel values changing during the course of the day. For the purposes of following the model strategies as closely as possible, you should limit your market actions to as near to the market close as possible.