Skip to main content

When non-fungible tokens (NFTs) were first introduced to the public in 2014, no one would have believed that the digital image of an ape could be worth well over a million dollars. However, as more mainstream attention turned towards digital assets in 2021, NFTs began to surge in popularity with their ability to represent an endless variety of objects: digital art, video game collectibles, music, etc. 

Launched in April 2021 with a mint price of 0.08 ETH (~$236), the Bored Ape Yacht Club (BAYC) collection now contains pieces priced at 3 million dollars.

The boom of the NFT marketplace also creates opportunities for financial criminals and introduces risks such as wash trading. Wash trading occurs when the same person or entity sells and buys back a “substantially identical” asset, and it was a tactic commonly used among stock market manipulators before the 1930s to create a false sense of demand and drive up stock prices.

Although wash trading with traditional assets had been made illegal with the Commodity Exchange Act in 1936, the NFT market is still new and largely unregulated due to ongoing active debate over which asset class NFTs belong to.

Thus, wash trading NFTs might escape government scrutiny and become attractive for multiple reasons:

Price inflation: Trading the same token repeatedly between multiple wallets may artificially inflate the token’s transaction volume to drive up its price. 

Money laundering: Criminals can move large sums of money between themselves under the cover of legitimate NFT transactions and volatile price changes of the crypto market. 

Tax loss harvesting: NFTs can be sold and repurchased at a loss to exploit loopholes in taxation law, enabling traders to claim a tax loss without losing ownership. 

The prevalence of wash trading currently poses a substantial threat to unsuspecting buyers in the NFT market. To tackle this issue and place more scrutiny on these financial criminals, AnChain.AI developed an algorithm to detect wash trading activity and calculate profits made by suspected wash traders. 

Detecting Wash Trades 

The algorithm goes through a collection’s transaction history and flags transactions where a wallet buys back the same token they had sold less than 30 days prior. It then identifies suspicious intermediate transactions between when the wallet first sells the token and when it makes the repurchase. To better visualize the token’s ownership history, our algorithm creates two directed graphs for each token: 

  • Token transfer graph displaying all transfers of a particular token including those where no money is exchanged 
  • Token sale graph displays all sales of a particular token where the token was exchanged for money 

Calculating Profit

The algorithm calculates profit made by wash traders in two different ways: price manipulation and sales:

Price manipulation profit: price of token exiting cycle – price of token entering cycle (how much the token’s price increased after the wash sale cycle)

Sales profit: token sale price starting cycle – previous token purchase price (how much the wash trader made from the original sale)

Both methods of calculation present a potentially important signal for risk categorization and predictive machine learning analytics.

Price Transaction History for Art Blocks ID 78000189 

Timestamp Source DestValue ($) 
2021-06-11 21:17:400x84dce06ea488377005cfb85900600118b4ad76950xb4968ae81fdd1bc3b8fbf832ffda61bc337f5118826.09
2021-06-12 22:23:250xb4968ae81fdd1bc3b8fbf832ffda61bc337f51180xdffc40a822b608daf51ada5104b11096d434d17d2,284.03
2021-07-31 19:37:120x9275776b9ac4c62ce29a3d3b62623c35961afa9e0xe1d29d0a39962a9a8d2a297ebe82e166f8b8ec1857,050.84
2021-08-15 22:21:340xe1d29d0a39962a9a8d2a297ebe82e166f8b8ec180xfe2f279d3679bac2d07cf46c93503410ef9ca448130,732.24
2021-08-23 18:04:09 0xfe2f279d3679bac2d07cf46c93503410ef9ca4480xe1d29d0a39962a9a8d2a297ebe82e166f8b8ec18402,418.76
0xe1d29d0a39962a9a8d2a297ebe82e166f8b8ec180xc77fa6c05b4e472feee7c0f9b20e70c5bf33a99b470,570.32

After running the algorithm to produce the graphic above and examining the price transaction history for Art Blocks Token 78000189, we found that the wallet KeyBoardMonkeyVault (address 0xe1d29d0a39962a9a8d2a297ebe82e166f8b8ec18) bought the token for $57K before the wash sale cycle and sold it for $471K 8 days later after the cycle, making a price manipulation profit of $413K. This represents one of the most substantial price manipulation profits made by a wallet among all 7 collections that we inspected.

Running the algorithm on the BAYC collection history––containing transactions on OpenSea from launching till June 2022––yields that around 0.275% of transactions, 0.414% of tokens, and 0.265% of wallets contain wash trading activity. Furthermore, the suspected wash trading wallets made a total price manipulation profit of $755K and a sales profit of $651K (detailed below).

Wash Sale Stats for BAYC:

# of Wash Sales72
% of Wash Sales0.275
# of Wash Traded Tokens36
% of Wash Traded Tokens0.414
# of Wash Trading Wallets34
% of Wash Trading Wallets0.265
Avg. # of Wash Sales in a Cycle2.194

Profit Stats for BAYC: 

PM Profit ($) Sales Profit ($) 
Max344,448.96 286,114.18
Average20,964.5318,094.01
Total 754,722.91651,384.41

Furthering our analysis to contain data from 7 popular NFT collections (Art Blocks, Azuki, BAYC, Decentraland, Doodles, Mutant Ape Yacht Club, Otherdeed), we found that an average of 0.14% of transactions, 0.11% of wallets, and 0.16% of tokens are suspected to contain with wash trade activity. These wash trades generate an overall price manipulation and sales profit of $930K and $1.1M. 

The large amounts of money being made by the suspicious wallets again underline how wash trading threatens the integrity of the NFT market. AnChain.AI’s algorithm offers a solution, designed to alert unsuspecting buyers of wash trading activity among certain tokens and wallets and is aligned with our goal to improve security in the web3 industry.

Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments