Looking to the largest accounts (~700 accounts; min of 10K protocol followers each).. community detection algorithm (Louvain) basically splits the social (sub)graph in two

Lighter nodes indicate lower FIDs.. thus, we can see that the (light blue) left side was "early" compared to the (dark blue) right side

Accounts with FIDs below 5K (left; 260 accounts) vs. accounts with FIDs above 300K (right; 150 accounts)

Nodes are sized by number of (following/followers) connections each account has with other accounts included in this data set (~700 accounts)

I like running, poker and oysters | Founder, Social Forensics (SocialForensics.com) | Berkeley & UPenn | Learning to listen

It's common for accounts to be connected to other accounts that joined around the same time

Take @matallo.eth (FID: 626), for example, whose position on the map is on the far left

That's because he's far more connected to accounts that were "early"

Accounts he is following (left) vs. followers (right)

Explorer. Code, design, and mountains. Building Fern https://fernhq.com Previously Shopify

Take @vlady (FID: 239,748), for example, whose position on the map is on the far right

That's because he's far more connected to accounts that were NOT "early"

Accounts he is following (left) vs. followers (right)

The position of my account is central in the map

That's because I am equally connected to accounts that were "early" and accounts that were NOT "early"

Accounts I am following (left) vs. followers (right)