For those of you unfamiliar with this pattern check out: http://www.surinotes.com/Tradestation/articles/suriDragon.pdf Its really just a way to trade double bottoms. The reason I chose this pattern to automate is because I have a few ideas on how I can automate the process of finding double bottoms. So I took that idea and expanded it to find "dragons".... My code looks to be finding the dragons pretty well. Granted it doesn't find them all, but that is not too important at this stage. As long as what it does find are dragons. Which for the most part it seems to do pretty well. Automating a strategy using the rules documented at the above link looks to be very promising, but does not yield an immensely profitable strategy.... This thread will be more "open and sharing" than most of my threads normally are. The aim of the thread is to try and improve the strategy by figuring out better exits, and by figuring out ways to filter out the dragons that have a low chance of success... I welcome anyone who trades this pattern, or wants to learn how to trade this pattern to join in on the discussion. The pattern seems to be a good fundamental base to build off of... but I think it can be improved a lot. NOTE: Although this thread will be more open than most of my threads, I will not be disclosing the algorithm for finding double bottoms or dragons. The thread will assume that everyone participating is finding the dragons manually and not through automated means.... The "openness" will revolve around exiting, stops, and filtering out the signals that have low chance of success.
Right off the bat I would say that the 2nd leg of the dragon be a higher low. You could do some testing on dragon's that have a lower low, exact double bottom and a higher low and compare them.
Nobody ever wants to hear about mucking with another indicator, but using some sort of acceleration/momentum indicator and looking for a divergence on the second leg might also lead to higher profitability...
Would not want to waste my time on that. That's already real old, just that there has been given a new name to it. The guy that initially called it a dragon has now giving up trading it. Cannot believe that they are trying to bring it as a good way for trading, but hey they have to sell magazine's don't they? Snake oil in overdrive: "find a profitable system in our magazine and it will only cost you a few dollars to buy it and it will make you all the riches in the world". As the Tui beer ad says here; "Yeah, right". Maria
Due to testing with range bars being soo slow with NT, I am only testing on a few months of data at this time... however, it appears as close to a true double bottom as possible seems to work best, anything where the 1st and 2nd leg are within 1 ES point of each other seems to do well...