NFT Twitter Visualized: 1,597 Stars in the Web3 Sky

We mapped the NFT Twitter community, and here’s what we learned about community structure and influence. Where do you fit in?

First and foremost, this is a work in progress. Inevitably there will be quite a few accounts that our Q1 2022 map of the NFT Twitter community is missing. While we do spend a fair amount of time immersed in NFT Twitter, we realize our first stab will be far from perfect. As a result, we would encourage feedback from the NFT Twitter community, as that will allow us to improve future versions of AnChain’s NFT Twitter community maps!

Background

We started with a set of a few hundred seed accounts and expanded from there to a map of Crypto Twitter that contains 5,092 of the most connected digital asset focused accounts (based on Following/Followers relationships). The green community, for example, contains 1,597 accounts that are (for the most part) comprised of accounts from the NFT Twitter community:

Our analysis moving forward will focus on the green 1,597 accounts above, i.e. the NFT Twitter community. The colored communities are determined algorithmically using the Louvain method for community detection, which effectively results in accounts that are Following similar accounts and that have similar Followers appearing within the same community. In other words, we are using Following/Followers relationships as a proxy for interests. And, in the case of Twitter — an interest driven social network — this approach to surfacing communities works particularly well.

Below are the 1,597 NFT Twitter accounts after reapplying visual clustering and community detection algorithms:

Community Identification

There were 4 algorithmically determined communities in what, moving forward, we’ll simply refer to as “NFT Twitter” (once again, this is by no means exhaustive, and we intend on expanding the map based on feedback from the community!):

The determination of communities is being driven entirely based on the Following and Followers of the 1,597 accounts in relation to the same for the other NFT Twitter accounts. In other words, it has nothing to do with what accounts are tweeting, which accounts they are engaging with (retweets, likes, replies), keywords appearing in account bio descriptions, or anything else, for that matter — just the (Following/Followers) interconnectivity between accounts.

Please think of the community-level categorizations to follow as “best fit” descriptions. They certainly won’t hold true for every single account within each respective community. As mentioned earlier, however, Twitter Following/Followers relationships are generally quite effective at identifying interest based communities. We will explain these categorizations in the context of NFT Twitter momentarily.

Not surprisingly, 42% of NFT Twitter accounts (676) contain the word “NFT” in their username, display name, and/or account bio descriptions. As can be seen, those accounts are distributed relatively evenly among the 4 communities: 

Enriching the NFT Twitter map with profile data from each Twitter account allows for quick (and visual!) assessment of the algorithmically determined communities.

Light Green: NFT Gaming/Cryptoart

Over the last few years NFTs have slowly made their way into the public sphere — and in the first half of 2021, they exploded into mainstream adoption.

As can be seen below, the light green community contains significantly fewer accounts that were created last year (2021) than the other communities:

Moreover, it contains a larger portion (26%) of accounts that have “crypto” in their bio descriptions when compared to other communities (16%, 11%, and 7%, respectively):

Broadly speaking, the light green community can be categorized as crypto native (compared to the NFT Twitter map overall, at least) — meaning accounts from this community generally were introduced to digital assets prior to NFTs going mainstream in 2021.

There is a mix of NFT gaming and cryptoart focused accounts within this community.

Take Neon District (@neondistrictRPG), for example, the product of Blockade Games, a venture backed startup that is an OG in the blockchain gaming space:

And then there is Axie Infinity, an NFT-based online game that was released in early 2018 (highlighted below are accounts that include “Axie” in their bio descriptions):

Included among those accounts is Axie Infinity co-founder, Aleksander Leonard Larsen (@Psycheout86):

The upper section of the light green community has more of a cryptoart than NFT gaming focus. Vastly oversimplifying, cryptoart was the precursor to NFTs making their way into the public sphere. Stated differently, crypto artists are the OG NFT artists.

Josie Bellini (@josiebellini), an artist and designer who has been creating cryptoart since 2017, would most certainly fit that bill:

Similarly we can see Osinachi (@osinachiart), a Nigerian visual artist known for using Microsoft Word as his medium, emerge as an account that is central within the upper section of the light green community:

Dark Green: NFT Artists

Both the light and dark green communities contain a larger portion of accounts (24% and 36%, respectively) that have “artist” in their bio descriptions when compared to the pink and blue communities (11% each, respectively):

The artists from the dark green community are generally newer to the NFT ecosystem than the OG crypto artists from the light green community.

Here’s an animation that cycles through the accounts being followed by three well known NFT artists (Bill Elis, FEWOCiOUS, and Beeple) from the dark green community:

Pink: Bored Ape Yacht Club (BAYC)

The pink community may be categorized as Bored Ape Yacht Club-centric, as can be seen by the large number of accounts (86) that have “BAYC” in their bio descriptions:

Here’s an animation that cycles through the accounts being followed by five accounts that are central to the pink community (@BoredApeYC, @andrwwang, @tropoFarmer, @BAYC5511, and @YatMuseum)

Blue: CryptoPunks

The blue community may be categorized as CryptoPunk-centric, as can be seen by the larger portion (19%) of accounts that have “punk” in their bio descriptions when compared to other communities (10% for BAYC and 3% each for the others, respectively):

Here’s an animation that cycles through the accounts being followed by five accounts that are central to the blue community (@punk6529, @punk7635, @punk4156, @punk607, and @punk3178):

Network Density

Our NFT Twitter network graph visualization contains 1,597 accounts and 475,745 relationships between said accounts. This means that, on average, each account in the map has nearly 300 connections (i.e. Following and/or Followers relationships) with other accounts in the map.

When isolating communities within the NFT Twitter map, here’s what network density looks like:

This level of connectivity is indicative of Twitter communities that are deeply intertwined.

One possible theory as to why NFT Twitter is such a densely connected network/community could be due to Twitter Spaces. Twitter’s social audio feature functions as a hub for NFT focused discussions.

The Spaces user experience allows for Following speakers/listeners with one tap during social audio sessions. Hence, the constant stream of NFT focused discussions on Spaces could be a contributing factor to NFT Twitter being one of the most deeply connected communities we have encountered.

Most Influential Accounts

Here are the 100 most influential NFT Twitter community accounts (based on eigenvector centrality algorithm; overall, not top 25 per community):

Takeaways/Conclusion

NFT Twitter emerging as such a deeply connected network doesn’t come as too much of a surprise. After all, community is the killer app for the metaverse.

Effectively, our NFT Twitter map is bifurcated into artist and collector sections.

Artists are split between the light and dark green communities (crypto and NFT artists, respectively), while collectors are split between pink and blue communities (BAYC and CryptoPunks, respectively).

Visually, the overlap within the artist (red) and collector (yellow) communities can clearly be seen:

And last, but certainly not least, AnChain.AI’s 2022 NFT Twitter map reflects significantly more overlap between NFT artists (dark green) and CryptoPunks (blue) when compared to NFT artists (dark green) and BAYC (pink).

If you enjoyed this post, you won’t want to miss our next one – where we will share insights from analyzing the on-chain behaviors of the wallets behind the most followed Twitter accounts with .eth names.

Be sure to subscribe to our newsletter and follow us on Twitter @AnChainAI to be notified once we publish our findings!