Cryptocurrency Data Science

Market Maturity, Developer Indexing and Token Utility

Flipside Crypto’s Cryptocurrency Data Science

We believe the Cryptocurrency market can be characterized by a combination of 3 traits: Market Maturity, Developer Indexing, and Token Utility.

Market Maturity 

Our crypto-asset maturity index relies on price and volume data inputs to characterize trading behavior on each potential investment. By modeling many variants of known algorithmic trading approaches (i.e.: like price momentum), we build models that execute optimized versions of those strategies and observe their ability to provide consistent returns across several market scenarios. High scores mean that many different strategies are likely currently being used (a more mature asset). Low scores indicate a risk for individual trades or groups of trades to move the market (pump-and-dump schemes), or assets that are not liquid enough for reliable regular trading.

Developer Indexing

Flipside’s Github Developer Index applies dimension-reduction techniques to data gathered from online code repositories such as Github. Because the vast majority of cryptocurrency projects run on open-source projects, we can use their data to reliably compare the level and trend of work that occurs on these projects across a variety of variables over time. Our analysis has shown that few tokens have seen a rise in price without a preceding rise in their indexed Developer Activity scores.

health index

Token Utility (Blockchain Nodes Tracker)

The Blockchain Nodes Tracker data streams directly from blockchains with an emphasis on understanding Token utility. We actively maintain nodes to ingest transaction, account (IP), and smart-contract details on a continuous basis. By combining straightforward summary details like numbers of transactions, ensuring a breadth of number of active accounts, and teasing out investment vs. utility activity, we can can characterize both how much a project’s tokens are traded, and to what end.

These variables cannot be explained explicitly, nor can they be modeled by one particular variable. We create sub-models that explain aspects of each, and then bring them together into a single model that accounts for the inter- and intra-relationships.