Groups and sorts markets by similarity based on the shape of their time series. This can be used for diversification and to identify anomalous versus generic markets.

The Grid tab shows the different clusters identified by the clustering algorithm sorted by their size. Finding different clusters in market data can help with diversification: Perhaps all your holdings are just in one cluster and therefore not very diversified based on their actual dynamics over time. The number of clusters can be defined by using the Clusters setting. The chart legend allows you to highlight different markets temporarily by hovering over them or permanently by clicking on them. Double-click on any chart to remove the permanent market selection. The color of the chart lines is determined by their distance from the most generic market.

You can see this in more detail by going to the List tab, where all the markets are sorted by their distance. The market with the highest distance (100%) is the most anomalous of all the markets analyzed. The market with the lowest distance (0%) is the most generic of all the markets analyzed.

As the crypto industry continues to mature, crypto assets may become less correlated and clustering may become more useful to identify different groups of assets.

Please note that sometimes there may not be enough data for a particular market to be included in the analysis, either because the selected Time Horizon is longer than the listing period, or because the data itself is simply too sparse.

Stablecoins are excluded from the analysis.