Who is trading on Trade.xyz?

Author: Arrakis

Compiled by: Jiahua, ChainCatcher

Introduction

When we wrote our first article on "Who is Trading on HIP-3," the attribution method was statistical.

We classified each wallet based on its trading behavior over the past three months: addresses primarily making orders (makers) were classified as market makers, addresses with high-frequency taking activity (takers) were classified as arbitrageurs, while addresses with low transaction rates and builder-tagged orders were considered retail investors.

While this revealed some interesting patterns in market structure, this classification is probabilistic, leaving about 70% of wallets in the market unclassified.

Arrakis @ArrakisFinance · April 8 Who is Trading on HIP-3? Analyzing order flow across seven HIP-3 perpetual contract markets on Hyperliquid, involving 175,000 wallets, $16.1 billion in trading volume, and the front-end interface routing it all.

This article replaces statistical inference with mechanical classification. Each order on Hyperliquid comes with a set of deterministic tags signed and published by the exchange itself: valid time (ALO, GTC, IOC, FrontendMarket), builder code, fill flag, holding time.

We used this order metadata to classify each wallet into one of four categories: retail, market maker, arbitrage bot, or airdrop farmer.

The next step was to identify the wallets behind these categories, extracting identity and trading behavior data from the Arkham and HyperTracker APIs.

The top 450 wallets contributed 78% of the total trading volume. Within this set, we identified several accounts associated with Polymarket, as well as institutions like Jump Crypto, Selini Capital, Wintermute, and Abraxas Capital.

Several clear patterns emerged from this two-step classification method. We will explore these in detail below.

Wallet Overview

We observed a 21-day period from March 10 to March 31, 2026.

During this time, four Trade.xyz markets (Crude Oil xyz:CL, Silver xyz:SILVER, Tesla xyz:TSLA, and xyz:XYZ100) recorded 79,622 unique participating wallets, with a total trading volume of $51.95 billion.

The 79,622 participating wallets during these 21 days broken down by trading volume. Market makers accounted for less than 0.5% of the total wallets but contributed 63% of the trading volume.

Classification of wallets by number rather than trading volume. The "airdrop farmer" category alone contains 35,091 wallets, nearly half of the identified total.

By wallet count, "airdrop farmers" are one of the largest categories, but by trading volume share, they are the smallest.

The 35,091 wallets account for 44.07% of the total wallets but generated only $400 million in trading volume during that time window, which is just 0.77% of the platform's total of $51.95 billion.

On Trade.xyz, nearly half of the active wallets generated trading volume of less than 1% of the market total.

Breaking them down by the markets they participated in revealed another surprising pattern.

Distribution of wallet counts by market. Due to optimal execution costs, xyz:CL attracted 99.3% of airdrop farmers.

Among the 35,091 wallets in the airdrop farmer category, 34,859 (99.3%) traded xyz:CL during that time window. The remaining 232 wallets were distributed among xyz:SILVER, xyz:TSLA, and xyz:XYZ100.

This pattern is highly consistent with airdrop farming behavior: each wallet engages in small back-to-back trades in both directions, generating trading volume without taking on price risk.

This strategy relies on tight execution costs and benefits from extremely low slippage. xyz:CL has the best depth among the four Trade.xyz markets, making it a natural breeding ground for such activities.

Another interesting observation is who is controlling these addresses. The on-chain tracking results presented later in the article point to the same Polymarket operator for 34,553 of the wash trading wallets.

During this time window, this entity alone accounted for 43.4% of all participating wallets on Trade.xyz.

At the other extreme of this classification is market-making behavior. 363 wallets (0.46% of active addresses) circulated $3.275 billion during the time window, accounting for 63% of all trading volume on Trade.xyz.

The remaining three categories occupy the middle ground: 522 SAT/HFT bots contributed $350 million (6.7%); 38,307 wallets classified as retail circulated $870 million (16.7%); and 5,339 wallets in the unclassified category contributed $661 million (12.7%).

This 12.7% trading volume in the unclassified category cannot be attributed to a defined strategy based solely on metadata.

There is reason to assume that a significant portion of this trading volume comes from retail investors placing limit orders through the Hyperliquid front end or placing market and limit orders through the Trade.xyz front end.

These channels did not attach explicit builder codes or dedicated TIF tags to such orders, rendering these transaction records invisible in metadata-driven classifications.

Weighted combinations of various TIF (Time in Force) types. Unsurprisingly, 98.5% of market maker orders are ALO, while 100% of arbitrage bot orders are IOC. In the unclassified category, 71.5% lean towards GTC, which is typical of limit orders manually placed by front-end users.

The combination of TIF also supports this hypothesis: 71.5% of orders aggregated into the unclassified category carry a GTC (Good Till Canceled) tag, typically used by front-end users placing limit orders.

Meet Themino, Themino, and Themino

In recent weeks, there has been extensive discussion in the industry around one question: Does the impressive number of users on Trade.xyz reflect real human participation, or is it inflated by wash trading activities anticipated from the TGE (Token Generation Event)?

While we cannot comment on the macro patterns of wash trading behavior across the entire platform, we discovered a noteworthy pattern after analyzing the transaction-level data from the four Trade.xyz markets in March.

Among the 34,602 wallets classified as airdrop farmers, 34,553 (99.9%) can be traced back to a single Polymarket identity named "Themino."

Themino Network. A Polymarket Arbitrum identity has derived 70 different linear links across 34,553 wash trading wallets.

Here’s how it works.

Hyperliquid's L1 has an internalTransfer function that allows USDC to be transferred between wallets, charging a flat fee of $1 regardless of the amount.

The operator of Themino utilizes this function to circulate an initial deposit across thousands of new wallets sequentially. Each wallet runs the same five steps in about 26 seconds:

1. Receives $X funds transferred from the previous wash trading wallet via internalTransfer, deducting a $1 HL transfer fee upon entry.
2. Transfers $14 to the xyz sub-account.
3. Executes two IOC orders on xyz:CL, one buy and one sell, generating two fills and creating some trading volume.
4. Transfers about $13.99 back to the main account (the few cents difference comes from execution slippage and trading fees).
5. Transfers $X minus $1 to the next wash trading wallet via internalTransfer.

The next wallet then repeats this sequence.

In the 34,510 internal transfer events of this operation, Themino incurred $34,510 in protocol fees, a pattern consistent with its trading history on Polymarket.

Themino also lost about $80,000 on Polymarket by betting "Will the U.S. strike Iran before February 28, 2026?" on the "No" option. The strike event ultimately occurred on February 28.

Builders

Hyperliquid attaches an identifier to orders routed through third-party front ends, allowing applications to charge custom front-end fees.

This identifier is the builder code, which is the most direct indicator of which interface a wallet is trading through.

Among the wallets participating in trading across these four markets, builders can be classified into three types.

Algorithmic Builders

These are products used by retail investors to maximize DEX trading volume to earn potential airdrop points.

Before the end of 2025, wash trading on perpetual contract DEXs meant wash trading or running directionless taker orders through algorithms, which is costly for participants and net negative for exchanges.

Retail market-making bots like Tread.fi, Planemo Trading, and Origami have replaced wash trading with valuable market-making behavior. Each order through these products is post-only, meaning wallets are adding liquidity to the order book rather than consuming it.

As Tread.fi CEO davidyjeong stated:

"Before retail market-making solutions emerged, wash trading on perpetual contract DEXs meant executing trades at the cost of fees, slippage, and the risk of being banned to artificially inflate trading volume.

We solved this problem by building a novel solution: bots only quote limit orders on both sides.

Users pay lower costs to wash volume and often profit from captured spreads, while the side effect is providing real optimal bid-ask liquidity to the market—exactly what HIP-3 stock perpetual contracts need during nights and weekends when traditional market makers do not quote.

This is a better way to wash volume and the reason why the HIP-3 market can achieve excellent execution depth today."

During periods when traditional market makers do not quote, these retail market-making bots contribute most significantly to the market.

CME's WTI crude oil futures close on Friday afternoons and reopen Sunday evenings, and stock perpetual contracts face the same overnight and weekend gaps. During these windows, retail market-making bots fill the top quotes of the order books for xyz:CL, xyz:TSLA, and similar markets.

Note that while we classified wallets routing orders through these algorithmic products as "airdrop farmers" in this analysis, their trading behavior and market impact are structurally different from wash trading activities.

Wallet Integrated Builders

Wallet integrated builders are perpetual contract trading interfaces embedded within consumer wallets.

Since early 2026, these integrated functions have collectively become one of the largest sources of retail order flow on HIP-3.

This group includes Phantom, MetaMask, Rabby, Rainbow, and OneKey. The median trading volume for each wallet ranges from $1,000 to $3,000, aligning with the order sizes of retail groups that prioritize access convenience over minimal front-end fees.

Standalone Applications

Standalone applications are independent perpetual contract front ends and integrated applications: trader-focused products that provide better order entry, chart analysis, position management, and execution tools for users who are not satisfied with wallet plugins and need dedicated workflows.

Compared to wallet integrated channels, this group has fewer wallets but higher trading volumes per wallet, fitting the core user profile that values functional depth over plug-and-play convenience.

Products include Insilico Terminal, Liquid Trading, Hypurrdash, BasedOneX, Dreamcash, Infinex, Pear Protocol, DeFi App, and pvp.trade.

0xVKTR, the growth lead behind Insilico Terminal, commented:

"At Insilico, we view the HIP-3 market as the next step in bringing real-world asset exposure natively into the crypto track. Traders want more than just another front end; they want fast execution, clear market access, and the ability to switch seamlessly between crypto assets and macro assets without leaving their existing workflow.

Trade.xyz is one of the clearest examples of this demand. The flow of funds routed through Insilico indicates that as long as the platform has depth, the product is useful, and the trading experience is designed for serious participants, on-chain perpetual contracts have a truly core user base."

Market Makers

The market-making landscape on Trade.xyz is highly concentrated. The top 5 market makers account for 50% of the total market making volume, the top 13 account for 80%, and the top 21 account for 90%. A few institutions handle the vast majority of market making orders on the platform.

Cumulative market-making trading volume share by wallet ranking. The top 5 institutions operate 50% of the market-making fund flow, the top 13 reach 80%, and the top 21 reach 90%.

The second-largest market maker is the most interesting wallet in the entire group.

0xc926ddba...98d3 completed $4.39 billion in trading volume with a 0.52% fill rate, a textbook maker characteristic. Arkham has marked this address as "Powell" on Polymarket.

One of the largest market makers on Trade.xyz is actually a Polymarket user operating a multi-market order book on HIP-3.

Other notable market-making institutions include:

Jump Crypto operates two wallets with a total trading volume of $3.15 billion. The funds come from 0xf584...d621, a Jump treasury wallet marked by Arkham clustering, holding over $160 million in diversified assets, including LINK, LIT, EIGEN, BNB, ETH, USDC, and USDT.

Selini Capital operates three wallets: two running pure limit order quotes via API (0x44a3e1...35dd, 0x76987c...4480) and one running pure aggressive taker orders (0x427be6...d1d9). Selini's total trading footprint reaches $1.03 billion.

Hyperliquid's own order flow tagging feature distinguishes Selini's market-making wallets from its high-frequency trading wallets. The same institution has layouts on both sides of the order book.

Wintermute operates a market-making wallet with a trading volume of $229.6 million (0xecb63caa...2b00), smaller than Jump or Selini. The funding source is OKX.

Top market makers on Trade.xyz (ranked by trading volume). Powell, Jump Crypto, Selini Capital, and Wintermute occupy shares of the order book with traceable identities.

Aside from named institutions, most market-making trading volume comes from wallets without identifiable entities but with clear funding sources: Kraken, Binance, Coinbase, UnitXYZ cross-chain bridges, or Hyperliquid native funds.

Their behavioral characteristics are those of market makers, but the operators behind the addresses have never interacted with tagged services that would allow Arkham to trace their identities.

The median order/fill ratio for market makers is 19.4, meaning for every fill, about 18 orders are placed and canceled as part of the bid-ask quotes.

The same top 5 market makers are simultaneously active across all four xyz markets, indicating they are multi-market quoting institutions rather than four independent businesses.

The market maker liquidation rate during the time window was 19.2%, nearly on par with the 20.4% for retail investors.

As quotes are filled, market makers on Trade.xyz take on directional risk: after a large sell order matches a buy quote, the institution passively holds a long position; after a large buy order hits a sell quote, the institution passively holds a short position.

During the severe fluctuations in crude oil in March, accumulated inventory positions moved rapidly in the opposite direction, outpacing the institutions' hedging capabilities, resulting in one-fifth of market makers experiencing at least one liquidation.

Statistical Arbitrageurs / High-Frequency Trading Bots

SAT (Statistical Arbitrage Takers) are the counterparty to market maker order books. These wallets run over 90% IOC (Immediate or Cancel) orders and are purely aggressive takers, specifically targeting market maker quotes.

Top SAT/HFT bots ranked by trading volume. The top 4 wallets account for 89% of the total SAT volume, with funding sources primarily concentrated in Bybit.

The top 4 SAT wallets account for $3.5 billion of the total $3.5 billion SAT volume, with a concentration of 89%.

Two of them run 100% IOC, meaning every order they place is immediate or cancel, with no intention to place limit orders.

Funding source clustering points to Bybit as the dominant SAT profile. Most of the top SAT trading volume can be traced back to wallets funded by Bybit, consistent with the characteristics of a single bot operating entity or a small community of bot operators.

Three SAT wallets have Polymarket accounts: loracles ($15.5 million in trading volume, $25.7 million cumulative profit on Hyperliquid), Conduit ($5.3 million), and ChadwickLongman ($3.7 million).

The reason Polymarket emerges here is the same as the reason it dominates the retail segment below—predictive market participants are the most common cross-platform identities on Trade.xyz.

The liquidation rate for SAT during the window was only 8.1%, about half that of market makers and retail investors. This group is the most stringent in risk control on the order book, hedging external positions and rarely triggering margin calls.

Retail Group

We conducted an in-depth study of the top 400 high-volume retail wallets routing orders through the Hyperliquid interface and builder-integrated front ends, excluding the algorithmic products mentioned earlier (Tread.fi, Planemo Trading, and the broader algorithmic builder group).

The top 400 high-volume retail wallets categorized by different attributes. Polymarket-tagged wallets dominate the identifiable identity share, contributing $1.63 billion in trading volume for this group.

Polymarket: Identified Polymarket accounts through Arkham's on-chain tagging system.

ENS / Social: ENS domain names or other identifiers from social graphs that cannot be traced back to Polymarket.

CEX Source: Anonymous wallets funded from centralized exchanges.

Cross-Chain Bridge Source: Anonymous wallets obtaining funds through cross-chain bridges.

Other: Self-custodied EOAs (Externally Owned Accounts), contract infrastructure, and activities from empty addresses.

The share of Polymarket is the most striking data point. Among the top 400 retail wallets, 94 are verifiable Polymarket users, accounting for 22% of the top retail trading volume ($1.63 billion).

This is the largest single identifiable category within the retail group.

Adding Polymarket market makers (Powell, $4.39 billion) and three Polymarket SATs ($24 million), Polymarket's total footprint on Trade.xyz is approximately $6 billion.

Selected Polymarket retail wallets among the top 15 trading volumes:

This overlap makes complete sense. Polymarket and Trade.xyz provide crypto-native exposure to real-world outcomes through different market structures (prediction markets vs. perpetual futures), and wallet-level data indicates that the same group of people is operating both ledgers simultaneously on the same EVM address.

ENS holders constitute another 26 retail wallets, totaling approximately $400 million.

Notable wallets include caydenb.eth ($33 million), eggnoodle.eth ($33 million), ethmerg.eth ($19 million), baitf1sh.eth ($16 million), and wanyekest69.eth ($6.8 million, with a cumulative PnL of $17.6 million).

The unidentified portion is categorized by funding source as follows:

CEX source retail wallets categorized by source exchange. Kraken dominates the unidentified CEX source group.

Cross-chain bridge source retail wallets categorized by source cross-chain bridge. Hyperunit and deBridge dominate this group, with Stargate and long-tail cross-chain bridges making up the remainder.

Kraken dominates the CEX source group. Hyperunit and deBridge dominate the cross-chain bridge source group, with Stargate and long-tail cross-chain bridges making up the remainder.

Thus, the true drivers of the Trade.xyz order book are actually three groups: Polymarket cross-border traders, independent traders funded by Kraken, and DeFi native users entering through Hyperunit or deBridge.

Integrating market makers, SATs, and retail groups into a unified view reveals a streamlined map of the Trade.xyz participant base (stripped of the wash trading layer).

Participant base map after stripping the wash trading layer. A few professional market makers and a few bot teams compete against each other, while the long tail consists of veteran Polymarket players and independent traders from CEX sources who place directional bets through the Hyperliquid interface.

A few professional market makers and a few bot teams quote against each other on the order book, along with the long tail of veteran Polymarket players and independent traders from CEX sources placing directional bets through the Hyperliquid interface.

This is the complete picture of the Trade.xyz order book.

Conclusion

Recent discussions around Trade.xyz have centered on one question: Is the participation on the platform real, or is it dominated by wash trading wallets inflating volume in anticipation of the TGE?

The analysis provides a layered answer.

Like any DeFi market before a TGE, Trade.xyz has a wash trading layer, such as a single operator running thousands of wallets in a relay farm. However, what the wash trading layer inflates is the number of wallets, not the dollar trading volume.

We found no evidence of large-scale wash trading operations built specifically to inflate dollar trading volume. The behaviors that appear to be wash trading in the data are mostly run by retail market-making bots: wallets place limit orders on both sides of the order book, increasing market depth rather than consuming it.

The existing real trading volume comes from identifiable ledgers. A significant portion of the top retail wallets carry Polymarket accounts, ENS records, or social graph identifiers, while the largest crypto liquidity institutions (Jump Crypto, Selini Capital, Wintermute) leave clear footprints alongside directional fund Abraxas Capital.

The retail group has an unusually large overlap with Polymarket, which itself is a product built around the same speculative demand.

The inflated number of wallets is a common phenomenon before a TGE. It does not extend to dollar trading volume, nor does it spread to identifiable groups routing these funds.

HyperTracker's API reveals the complete depth of the Hyperliquid market, which is not accessible elsewhere.

We segmented millions of wallets into different groups and tracked their behavior over time, creating one of the richest datasets for analyzing and backtesting real trader behavior.

Most importantly, we reconstructed trades from raw transaction data in near real-time, providing real-time visibility into positions, open orders, and the evolving market structure.

While the application showcases the possibilities, the API is the hidden layer that supports it all. For funds, algorithmic teams, and builders seeking a true edge, this is a comprehensive intelligent tech stack."

--- CoinMarketMan and HyperTracker co-founder djkanzen