The Solana NFT ecosystem has proven to be very welcoming of new technologies and approaches that enable trading to be more efficient and potentially boost profitability for the involved parties. The recently launched Hadeswap is one of the most contrasting platforms to the classic marketplace MagicEden, but its AMM model, 0% platform fees and 0% creator fees (royalties) raised concerns over how it might be used for wash trades. However, ME has also recently moved to making creator fees optional, which opens the way for similar behavior on its platform.

In this report we will have a look at the wash trading on both platforms and compare this activity over the last month to see whether either of them is more fertile ground for practicing the black art of boosting collection trading volumes.

Taking the Investopedia definition of wash trading:

Wash trading is a process whereby a trader buys and sells a [NFT] for the express purpose of feeding misleading information to the market. In some situations, wash trades are executed by a trader and a broker who are colluding with each other, and other times wash trades are executed by investors acting as both the buyer and the seller of the security.

In this report we are dealing with two types of marketplaces, each of them requiring different strategies in order to wash trade. For Hadeswap, the user always sells/buys to/from the collection pool, so there is no need for another address. They can simply sell, buy, repeat. For MagicEden, both a seller and a buyer account are needed to make a trade. Sometimes users try to obfuscate wash trading by transferring the NFT to other addresses or routing it through other marketplaces before ending up back in the initial wallet to repeat the cycle.

However, the end result of each method and the most certain way of identifying colluding users in the context of on-chain anonymity is by the fact that they each end up buying and selling the same mint more than once in a relatively short time frame.

In this report a default wash trade time interval of 1 day is used, meaning that repeat trades (sales/purchases) in the span of 24 hours from the previous transaction with the same mint by the same user address will be counted as being a wash trade.

The interval can be experimented with in the live dashboard by modifying the wash_days parameter.

All of the data and queries can further be explored and interacted with in the FlipsideCrypto dashboard below:

☝️ The dashboard can be updated on-demand by pressing the refresh button 🔄 in the top right corner of the page.

I. Washtrade Volume

The amount of wash trading on Magic Eden far exceeds that on Hadeswap. Over the last few weeks we can see average wash trading volumes of over $600k on ME and riding an upward trend since mid October. This current volume of wash trades oftentimes exceeds the entire trading volume going through Hadeswap. For its part, we see much lower wash trading volumes of under $5k over the last week and following a strong downtrend.

So if Hadeswap has 0% platform fees and 0% creator fees (royalties), why is it not the premiere platform to wash trade on? There are a few reasons.

First, it lacks the extremely large catalogue of collections MagicEden has so it naturally draws a lot fewer users (or potential victims).

Secondly, while those two fees are indeed set to 0% (for the time being), the platform is not particularly designed to facilitate wash trading or allow users to purchase at lower prices than anywhere else. Instead it is designed to maximize profitability for Liquidity Providers who can set their own swap fees (spreads) and a bonding curve for the price to follow. These LP set parameters often times result in a higher cost for users who exclusively trade on the platform.

Third, when ME users choose to pay 0% royalties, that amount is subtracted from the price of the item, leaving traders with just the 2% platform fee. This makes wash trading on ME more manageable and predictable in terms of cost, plus it benefits from a larger user base.

A direct volume comparison is not fair as HS is a recently launched platform and ME the long-established market leader.

Comparing the share of volume by its type, we see however that ME still has more of its daily volumes coming from wash trades. This got accentuated on October 15, the day it introduced optional royalties and when the platform had more than half of its volume marked as wash trading. Since then the share decreased a little, but without a doubt it contributed to the monthly wash trading ratio of 21% we can see in the pie chart.

The average value of wash traded NFTs has remained somewhat constant throughout the observed period, between $30 and $60.

Hadeswap on the other hand has very daily shares of wash trade volumes, oftentimes representing less than 1%. Since its launch almost a month ago on September 25, only 2.33% of its volume has been marked as wash trading.

The average value of wash traded NFTs has seen a lot of variability but seems to have been decreasing over time, currently in a range of $50 to $150.

II. Transactions

The transactions section of the dashboard tells a very similar story. MagicEden has a lot more wash trading on its platform by sheer size and most of the time the number of wash trading transactions alone is higher than the overall transaction count on Hadeswap.

But once again we see that wash trades make up a larger proportion of MagicEden’s activity, with almost a third of transaction being wash trades. In the case of Hadeswap, a little over 2.5% of the transactions since launch have been wash trades.

On average, 10 to 16 wash trades are made on Hadeswap every day over the last week and the trend line points downwards. On MagicEden there are around 20k wash trade transactions every day over the last week. This count does not seem to have changed too much following the Oct 15 optional royalties release.

III. Wash Traders

There are naturally some users who push wash trading to the limit and in the charts below we can see who the top wash trading users are based on their wash trade volume.

On Hadeswap, two accounts managed to surpass all other users in just 5 transactions where they managed to user the platform to trade mints back and forth with each other. Because both the sale and purchase of the mint were executed in the same transaction, the $31.5k volume is attributed to both addresses. The accounts were exclusively used for these 5 transactions.

For the rest of wash traders we see they weren’t created with the sole purpose to wash trade, but most of their activity and volume is nonetheless marked as such. Their activity is made up of regular interactions with the platform, unlike the unusual transactions executed by the top two addresses.

On MagicEden we see the top wash traders being responsible for up to 10 times more volume, with the first spot having generated over $444k wash trading volume with 390 transactions over the last month. The users here are split between being used primarily to wash trade or just casually doing it despite the large volumes.

Below we can see a distribution of wash traders based on the volume they have generated over the last month. Magic Eden wash traders are far more numerous and wash trade much greater volumes than those on Hadeswap.

IV. Wash Traded NFT Collections

The most wash traded collections on Hadeswap over the last month are Degenerate Ape Academy, followed by ABC and Critters Cult.

For Degen Apes, the wash trading volume represents almost 30% of the total volume they’ve had on the platform over the observed period, a lot higher than all the other projects in the top 10. Other collections with high wash trade volume share are Pawnshop Gnomies and FRAKT with 9.8% each, Netrunner with 8.5% and SoDead with 7%.

The most wash traded collections on MagicEden outsize again Hadeswap, with the top collections being Critters Club, y00ts and DeGods. Critters Club had a wash trading volume of close to $1M over the past month, but even so it represents just 15% of the total volume the collection had on the marketplace.

While most other collections in the top 10 have a similar share of volume, three collections stand out with almost all their volume being marked as wash trading. Those collections are NFTrees Solana with 92%, Copetown with 96% and Greedy Giraffes with close to 100% of the volume on ME being just wash trade.

What is even more shocking in the case of these three intensely wash traded collections is that they achieved these volumes using just a handful of NFTs out of the entire collection. NFTrees, a collection of 10k items, has a washtrading volume of $300k using just 622 distinct mints.

Conclusion

Hadeswap not only has less wash trading by volume and transactions, but they represent a smaller proportion of the overall activity on the platform and the trend continues to head downward. On MagicEden however, wash trade volume got a boost as soon as optional royalties were introduced on Oct 15. Even before that, the marketplace had a lot of wash trading going on and while transaction counts seem to slow down, wash trade volumes are holding up.

Another way they differ is by the type of users that wash trade and their budgets. Magic Eden wash traders are far more numerous and spend much more on wash trading.

The same difference in size can be observed when we look at the most wash traded collections. While the most wash traded collection on Hadeswap, Degenerate Ape Academy, barely goes over $60k in volume, the most wash traded collection on MagicEden, Critters Club, almost reached $1m.

Although we might be tempted to think that 0% fee platforms might be the most inviting for wash trading, there are more aspects that come into play such as spreads, liquidity and user base. Because wash trading is a short lived illusion, those who try it benefit from larger marketplaces with a diverse catalogue of collections, larger audience and preferably no additional fees. As it turns out, Magic Eden is just the place for that with over 90% of the market share on Solana by pretty much all metrics and with recently introduced optional royalties.

Edit: After a data refresh on Oct 24, new transactions were included in the solana.core.fact_nft_sales tables used for the Hadeswap transactions which resulted in considerable additional volume for the date of Oct 9. Most of the volume is associated with unlabeled/unknown collections but a brief analysis links many of the transactions to the LunarNFT project.

Although the volume and number of transactions are significant for that particular date, they do not have a big impact over the rest of the analysis and observations.