Crypto-ML’s trading system leverages multiple machine learning trading strategies to deliver trades to our Trader and Auto Trader customers. In this guide, you’ll find information on how each trade strategy functions.
Overview of Trade Strategies
Crypto-ML currently offers four trading and investment strategies:
|Hold||Keeps a designated percentage of your portfolio held in cryptocurrency.|
|CML-T||Applies deep learning to technical analysis data in order to generate a high-probability rules-based system.|
|CML-A||Measures the predictions from our neural networks to find anomalies in the market that indicate market manipulation that should be followed.|
|CML-X||Human-curated trades from strategies still in the research-and-development phase.|
Except for Hold, these strategies all seek to compound your money over time by capturing numerous small wins. Our machine learning clearly proves out this is the best long-term approach. However, no single strategy works in all conditions. That is why it is advantageous to run multiple strategies.
Important: This page is being updated and does not yet reflect the new trade strategies, including Hold, CML-T, CML-A, and CML-X.
CML-A: Anomaly Detection
CML-A trading strategy issues trades by finding anomalies in our neural network price predictions. This ideally allows us to identify and trade with unexpected events, market manipulation, and other exceptional scenarios.
Our machine learning seeks to predict market movements based on past patterns. In other words, it may say “when the market looks like this, 96% of the time, price moves up.”
The reality is machine learning will likely never predict 100% of market movements. There is always going to be a certain amount of chaos. And there will always be a certain amount of anomalous behavior, whether it be market manipulation, unexpected news, or some other exceptional event.
If one of our neural networks predictions are right 94% of the time, but we know that other 6% is highly interesting, then the CML-A strategy is our solution.
CML-A exploits the known limitations of machine learning and statistics to identify unusual price movements as they are unfolding.
The name “Anomaly Detection” comes from an understanding of the way it operates. Here are some examples:
- When the market patterns look like they do now, 97% of the time, price goes up by a small amount. However, now it is going up by a large amount. Current behavior does not match past patterns. This is a bullish anomaly. Something unusual is causing the market to go up.
- Likewise, if the market normally would go down but instead goes up, this is a bullish anomaly as well. Instead of pulling back like normal, some unusual force is pushing price up.
- These situations can also be viewed through a bearish anomaly lens. When the market patterns look normally result in a sideways price movement, but instead price is dropping quickly, there must be something causing this unusual movement.
As you can see in these examples, the market is behaving abnormally. It is likely being manipulated by some unusual activity or force. It is this force we want to identify and follow.
Strengths of CML-A
Whereas machine learning predictions may miss out on strong rallies and poorly handle strong drops (because these are exceptional conditions), CML-A will better capture rallies and better avoid drops.
CML-A has the ability to generate large gains and constantly trade through major upswings.
Weaknesses of CML-A
Under normal, non-extreme market conditions, CML-A may deliver choppy trades that result in small losses.
► Read the complete CML-A Overview
CML-T: Technical Deep Learning
CML-T performs a deep, novel, technical assessment of the market across numerous layers of market data. It utilizes this information to target extremely high-probability trades. These are trades that match patterns of highly successful trades in the past.
CML-T performs deep technical assessments of the markets across multiple layers to deliver high-probability trades.
CML-T performs complex equations but is not basing decisions on the output from our neural networks. Despite this, there are several areas machine learning contributes to this solution:
- Deep learning uncovers key patterns and relationships in the market.
- Optimization algorithms tune hundreds of parameters.
These findings and parameters are then curated into the CML-T solution.
Strengths of CML-T
CML-T seeks to issue very high probability trades. Its goal is to rapidly compound your trading funds.
Since it doesn’t rely on neural networks for buy and sell decisions, it can execute trades extremely quickly with little lag. This reduces trade slippage.
Weaknesses of CML-T
There may be extended periods that CML-T stays dormant. If it identifies unfavorable market conditions, it will avoid trading altogether.
► Read the complete CML-T Overview
CML-X: Experimental, Human-Curated Strategies
As a platform, Crypto-ML is constantly evolving. We have team members dedicated to advancing our strategies through research and development.
These advances may become available to Crypto-ML customers in one of two ways:
- As an entirely new strategy.
- As an enhancement to existing strategies.
Prior to release, all changes go through extensive statistical testing. Once that is passed, the changes run against live data to evaluate real-world performance. At any given time, Crypto-ML has multiple concepts running in the background at this stage.
Oftentimes, as we are monitoring these evaluations, certain clear patterns arise across multiple tests indicating opportunities to buy or sell. When this arise, we may issue the corresponding signal to the CML-X trading strategy channel.
It’s important to note that all CML-X signals will be curated by a human. Since these signals are being generated by systems in an evaluation phase, we want a degree of oversight to be applied.
Being human-curated, the signals will likely be:
- Less frequent in nature.
- May stay open longer than other strategies.
CML-X represents the latest in technology from Crypto-ML. As such, you’re receiving signals that may be more sophisticated in their understanding of the markets. In addition, these signals will typically be building on top of the knowledge gained from our existing platform.
Since CML-X is experimental in nature, there may be a higher likelihood of error or unanticipated behavior.
► Read the complete CML-X Overview