There are software releases, and then there are software releases that push the envelope…
This past week we released our most significant update since we began to roll out the platform earlier this year. We’ve made major improvements to virtually every aspect of the WaveBasis system and released several powerful new features. Current users may have already noticed some of the new features appearing in their workspaces.
The latest round of changes makes WaveBasis not only even easier to use, but also more effective in helping to identify low-risk, high-reward trading opportunities.
This first of a two-part update focuses more on “how the sausage is made”, so it’s a bit more technical than the second part. Part 2 is more practical, as it focuses on the new application features that are included in the latest release. We love trading, but we also love the geeky data science part of our work. So I hope you enjoy hearing a bit about it in this first part.
The most significant areas of improvement included in this platform update concern the Elliott Wave pattern detection engine itself, which we’ve affectionately (albeit boringly) been referring to internally as “V2”. In simple terms, the automatic wave count engine does its work in 2 main phases: detection and analysis. In this release we’ve focused on both of these phases. We’ve enhanced the capabilities of the pattern detection phase, and we’ve made the post-detection statistical analysis phase considerably more robust.
The enhancements we’ve made have resulted in a great increase in the statistical significance of the wave counts that V2 produces. Previous results were great, but V2 performs even better.
For some context: Since there are usually at least a few different ways to count a given section of a chart, successful application of the Elliott Wave Theory relies heavily on determining which wave count best captures a market’s true pattern context. This is measured in terms of probabilities, and each candidate wave count can be assigned a certain probability. Further, trading based on the highest probability wave counts is the only way that one can consistently and reliably identify good entry setups and exit points using the Elliott Wave approach. Stronger wave count probabilities means stronger trades and better profitability.
The latest incarnation of the computational engine identifies the highest probability wave counts better than ever before. It’s another new day and a new breed, and we’re extremely excited and proud about the series of technical breakthroughs we’ve experienced.
Additionally, the consistency of automatic wave counts has increased considerably. Counts are now more consistent both over time and within a given individual wave count analysis, offering higher confidence in any given wave count. This also makes it even easier and more intuitive to track a wave count over time. Improved consistency also notably impacts the Alternate Wave Count gadget, making it even more useful, which I’ll discuss further in an upcoming post.
While we were under the hood we also performed several optimizations, consequently improving the speed of automatic wave counts across the board. In some cases, however, the speed improvements are quite dramatic, as noted below.
Since most of the changes to the computational engine have been made behind the scenes and can manifest in quite subtle ways, many existing users might not notice any differences. However, depending on your trading style, you might notice some significant improvements to your automatic wave counts. Let me explain…
Better Long Term Wave Counts
The previous incarnation of the engine tended to be better optimized for shorter term traders, and feedback from WaveBasis users helped confirm this. So, although wave counts have improved for all time frames with V2, including shorter ones, scalpers might not notice much of a difference. The latest upgrades may be most noticeable to longer term traders. Daily, Weekly, and Monthly automatic wave counts now enjoy the same level of pattern recognition optimization and high-probability as sub-daily wave counts did previously. The playing field has been leveled.
Stronger Forex Wave Counts
We’ve also added detection components that have been specifically designed for Forex Elliott Wave pattern analysis. Consequently, Forex traders might also notice some significant improvements. Our ongoing research has consistently revealed that currency data has unique properties from an Elliott Wave analysis perspective when contrasted with other asset classes. Of course though, the fundamental Elliott Wave patterns themselves are exactly the same across all asset classes. However, the “shape” and “personality” of the underlying Forex time-series data presents special challenges for automatic wave counting, aside from the fact that they are 24-hour markets.
Although the unique properties of Forex price data can’t generally be observed by the naked eye, the results of our research, with the help of a collection of special-purpose algorithms, motivated us to allow the engine to treat Forex wave count analyses slightly differently than other markets. This has the overall effect of coaxing the best possible (highest probability) wave counts to the surface of each individual currency pair wave count analysis better than ever before.
Faster Automatic Wave Counts
Forex wave counts are also where we found the biggest speed gains. So, depending on the currency pair and time period, wave counts can be as much as 10 times faster! In our measurements, however, all wave counts across all asset classes should be at least twice as fast as they were previously!
An Important Practical Change
We also made a change to the fundamental approach to initiating automatic wave counts that makes counting waves easier for all users.
One of the things that folks tend to struggle with most is deciding where to begin a particular wave count analysis. This is also one of the questions that our support team hears most often. So, we’ve adjusted the wave count initiation approach to help to solve this particular problem.
In the latest release, in addition to new ways to define and execute an automatic wave count as described in Part 2 of this update, automatic wave counts now provide Automatic Pivot Detection. Once you’ve selected a date range for an automatic wave count, WaveBasis will automatically determine the most significant market pivot point within the selected date range, and use that as the beginning of the wave count.
This feature completely removes the guesswork of figuring out where to start a wave count, and makes it especially easy for beginners to get started. As long as the date range that you select includes a sensible pivot point, WaveBasis will do the rest!
This enhancement also helps make sure that resulting wave counts are optimally relevant for a given date range. I’ll talk more about this feature and several others in the second part of this update, so be sure to check out Part 2.
Taken together, the latest features and pattern detection enhancements advance the science and craft of Elliott Wave analysis forward yet another major step. This allows WaveBasis users to further bridge the gap between Elliott Wave Theory and the practical application of the Elliott Wave Principle.