In our recent webinar, we showed our approach to building trading systems. We went over a five step process starting with observation and hypothesis, transitioning to testing and verification, and finally, monitoring system performance. Following these steps is key for creating creating and maintaining a robust trading strategy. Check out our webinar recording here:
Video Outline:
As mentioned, we start with an observation. This will be a pattern that we’ve seen repeating itself over and over again (or at least we think so).
Based on this we establish a hypothesis for that observation that can be objectively tested. From there we go through a preliminary test that leads to verification or contradiction of the hypothesis. Of course, if verified, we‘ll want to monitor performance moving forward, making sure that the hypothesis stays valid.
Working with Bloodhound:
We first looked at some chart examples and methods for testing, and then showed at how we can implement this with a NinjaTrader add-on called Bloodhound (from SharkIndicators).
What we’ll do is to transfer an observation we’ve made into a computer testable hypothesis. To do that we‘ll use Bloodhound to specify our trading rules in a definitive and objective form. You can build whatever rule for whichever indicator with Bloodhound, and then associate each condition with different solvers. We will use two types of solvers to bring in the LizardTrader pivot S/R levels, namely the indicator comparison and the support resistance solvers. We then expand on this by adding reversal setups from the data series that come with the LizardTrader Auction Bars. Finally, we show how to optimize with additional function nodes to manipulate and filter the signals.
Of course, the whole purpose of this exercise is to test and establish that the connection between reversal patterns that plot at key S/R levels, and later price movement, is not a coincidence.