A view of the Mastercoin system

This is an attempt to get a birds-eye view of the Mastercoin system. By collecting all addresses and for each address build up a list of all outgoing transactions I have been able to build a few graphs which I will show here. To render these graphs the Excel template NodeXL was used and I will post a link to the Excel file at the end if anyone wants to play with this further. In March 2014 there were roughly 2700 unique Mastercoin addresses, out of which 2288 had a positive balance in MSC. In all graphs a vertex (circle) is an address, and an edge (arrow) is a transaction from one vertex to another.  

The first graph shows a raw overview containing all collected data (click for a larger image). As you can see the graph is directed in the direction of sent payments.

As expected there are many outlier vertices with only one edge pointed to, which is an address that only have received payments from a single address. There are also vertices which appear to be more central, connected by a large number of edges. If you look closely you will see a few arrows that are perfect circles. This is an address sending a payment to itself.

The vertices connected by many edges are obviously of some significance so I'll focus on them next. The next four graphs are sub-graphs of the the four vertices with the highest number of outgoing edges, i.e. addresses which have sent many transactions. 

The look very similar and resemble a hub and spoke pattern where one vertex in the center connect to many but the connected vertices does not connect to any or a few. Edges, if they exist, between vertices in the spoke are included and examples are seen primarily in the two graphs on the left.

Next I'll create four sub graphs for the four vertices with the highest amount sent in MSC. Out of these four vertices only one is included the the four above. (What is special about this address?)

Next, I'll start with the full graph in the beginning but will filter the graph by outgoing degree, i.e. sent transactions and raise this number gradually. The size of the vertices are also scaled where a higher number of transactions shows a larger circle and its presented in a grid with equal spacing between vertices.

Outgoing degree 1 and higher:

Outgoing degree 4 and higher:

Outgoing degree 10 and higher:

Outgoing degree 20 and higher:

To finish this I'll take a look at some metrics for graph theory:

Vertices*	2288
Total edges	3613		Number of payments sent
Unique Edges	2530		Payment only occurred once
Self-loops	4		Payment with same sender/receiver
Connected Components	5		Islands with no connection via payments
Vertices in largest component	2279
Diameter	12		Largest shortest path between vertices (how far have money travelled)
Max out-degree	245		Sent payments from address
Max in-degree	150		Received payments to address
Max (Eigenvector) centrality	0.034		Payments between an address with large number of payments to another with large number of payments
Avg (Closeness) centrality	0.003		Distance from one address to all other addresses. If equal to 1, the will be for every address one payment to all other addresses
Avg Clustering	0.017		Payments between receivers of of a payment
Reciprocal payments	232		Receiver sent payment back to sender

 * as mentioned in first paragraph