2% of users contribute 90% of trading volume: The true profile of Polymarket

Original author: sealaunch intelligence

Original translation: Chopper, Foresight News

Most reports about Polymarket only scratch the surface: milestones in trading volume, user growth, transaction counts, open positions, but never delve into who is actually trading behind these numbers. This article classifies all active wallets based on trading frequency and trading volume, outlining the true user profile structure of Polymarket.

The vast majority of trading volume on Polymarket is contributed by a small group of algorithmic traders and high-frequency trading entities; the massive low-frequency retail investors have almost no overlap with this group of professional traders. Recognizing the differences between these two groups directly influences platform fee structures, product prioritization, and strategic market category layouts.

Note: All data in this article comes from the Dune data dashboard, covering nearly three months of comprehensive wallet-level behavior; user profiles are defined by cross-referencing trading frequency tiers (T1–T7) and trading amount tiers (V1–V7), with amounts measured in US dollars.

User Trading Frequency and Volume Distribution

Trading frequency exhibits a typical log-normal distribution decay pattern. The largest user group, accounting for 32% of all users, traded between 2 and 10 times during the entire study period. Including those with 11 to 50 trades, this group makes up nearly two-thirds of the total user base. These users typically participate during elections, sports events, or major macroeconomic events, betting small amounts.

Trading frequency distribution chart

The distribution of trading volume, however, is markedly different. Although transaction frequency sharply declines from the left, the histogram of trading volume on a logarithmic scale forms a bell curve, peaking at about $600 to $3,000 per user. This indicates that typical active users have trading amounts in the four-digit range, but the tail end starting from $25,000 includes fewer users, yet they account for the majority of the platform’s total trading volume.

Trading volume distribution chart

These two histograms together reveal a structural split: one part consists of low-frequency participants; the other part consists of high-volume traders whose footprints are nearly invisible on the user chart but dominate the trading volume chart.

User proportion & volume concentration matrix is more intuitive: user dimension concentrates in the low-frequency small amount range, while volume dimension is completely reversed

How to Build a User Profile System

Relying solely on frequency or volume to categorize users overlooks the correlation logic between the two. For example, making 500 trades totaling $50 is entirely different from making 500 trades totaling $5 million. We classify each wallet based on these two dimensions.

First, each wallet is assigned to a trading frequency tier: from T1 (single trade) to T7 (more than 10,000 trades). Then, it is assigned to a trading volume tier: from V1 (total trading volume under $100) to V7 (over $2 million). The intersection of these two dimensions results in seven distinct user profiles, each representing a different type of participant.

• P1 Single-Trade User: Only 1 trade, total less than $100, a one-time trial to experience the platform
• P2 Low-Activity Retail Investor: 2–10 trades, total under $1,000, purely event-driven casual participants
• P3 Moderate Participant: 11–200 trades, total $1,000–$10,000, repeatedly entering but without systematic trading logic
• P4 High-Depth Retail Investor: 201–1,000 trades, total $10,000–$100,000, actively participating with stability but not at institutional levels
• P5 Low-Frequency High-Net-Worth Trader: fewer than 50 trades, with single large transactions exceeding $100,000, selectively seizing opportunities and holding concentrated positions
• P6 High-Frequency Professional Trader: over 200 trades, total over $100,000, employing algorithmic strategies and representing institutional traders
• P7 High-Frequency Small-Amount Player: over 200 trades, total less than $10,000, highly active but with limited capital

2% of Users Account for Nearly 90% of Trading Volume

The P2 low-activity retail group numbers as many as 849,000 users, accounting for 69% of all users; the P6 high-frequency high-capital group has only 27,000 users, about 2%.

However, during the analysis period, the P6 group generated a total trading volume of up to $39 billion. This is the most extreme manifestation of the Pareto principle: it’s not the conventional 80/20, but rather 2% of users supporting nearly 90% of the trading volume.

User profile summary table: Seven major user types derived from the intersection of trading frequency and trading scale

Median user counts, transaction counts, and transaction amounts: three data sets showing distinctly different user distribution characteristics

The user growth chart and the trading volume growth chart almost describe entirely different user groups. Platforms focused on user growth versus those focused on trading volume growth will have fundamentally different product strategies.

Category Preferences of Different User Profiles

Sports and cryptocurrencies are the two largest trading categories on Polymarket, accounting for 42% and 31% of total trading volume respectively, with significant differences in user demographics behind them.

Trading volume share by user profile and category

The proportion of high-frequency high-capital (P6) traders in the cryptocurrency market is significantly higher than in the overall user base, consistent with algorithmic trading models. These participants are not casual bettors but utilize systematic strategies for crypto trading. Their high trading volume and frequency indicate automated execution rather than subjective judgment.

Transaction count share by user profile and category

While sports betting is also dominated by high-frequency, high-capital (P6) trading volume, the proportion of medium (P3) and high (P4) participation users is higher than in the crypto category. Sports betting involves both institutional algorithmic funds and many experienced manual judgment players who rely on subjective analysis to place bets, rather than high-frequency machine-driven trading.

User proportion across profiles and categories: user distribution is opposite to trading volume and transaction count

Political users account for the highest proportion at 19%, but their numbers are relatively evenly distributed across user groups. Low-participation users (P2) are most prevalent among political users; compared to other categories, these are typically one-time retail investors driven by specific events, registering accounts solely to participate in election betting.

The economic and financial sectors attract a disproportionate number of low-frequency high-capital (P5) participants, meaning they trade infrequently but with large single transactions, investing significant capital into macroeconomic outcomes, with relatively few trades.

The categories on the platform directly influence the user groups attracted and impact liquidity depth, user retention, and fee tolerance.

A new cryptocurrency market will attract algorithmic and high-frequency traders; a new political market will attract event-driven participants who may never return after the event ends. More specialized markets, such as binary options or structured outcome markets, may further attract high-frequency high-capital (P6) users, and these systematic traders have already dominated the crypto space. If the goal is trading volume, build for the P6 user group; if the goal is user growth and brand influence, target the P2 user group. These two objectives require very different category strategies.

Implications for Fee Models

User segmentation directly influences the fee structure for predictive markets.

A fixed per-transaction fee model would excessively suppress P6 high-frequency high-capital and P7 high-frequency small-amount groups; yet these groups are precisely the ones supporting the liquidity foundation of the platform.

The value of differentiated fee rates across categories lies here, and Polymarket’s current fee system reflects this logic:

• Cryptocurrency segment: effective fee rate of 1.80%
• Sports segment: 0.75%
• Political & Financial segments: 1.00%
• Geopolitical segment: zero fees throughout

This fee structure is not arbitrarily set but is carefully aligned with the user demographics and trading habits of each category. The crypto sector is filled with P6 algorithmic professional funds, capable of bearing high fees without disrupting liquidity; the political sector mainly attracts low-barrier retail investors, requiring lower friction costs to maintain retention. Designing fee models without considering user profiles is essentially blind trial and error.

Core Conclusions

• The P6 high-frequency high-capital group comprises only 2% of users but generates 88% of the platform’s trading volume;
• Fee policies that harm P6 interests will severely undermine the platform’s foundation;
• 69% of users are low-frequency small retail investors, driven solely by hot events;
• Crypto trading is highly concentrated among algorithmic high-frequency funds, while sports participants are more diverse;
• The average user completes only 12 trades in 90 days, with a median total investment of $224;
• Expanding into new categories must be anchored to targeted user profiles, not just chasing trending topics.

Conclusion

If trading volume is concentrated in a small high-frequency core, why does Polymarket position itself as a retail product? Professional algorithmic funds support most of the flow, yet the product experience, marketing strategies, and category layouts always cater to ordinary retail investors.

Part of the answer may lie in structural factors. The proliferation of intelligent agent frameworks, Telegram bots, and no-code tools enables retail investors to easily automate trading. If retail investors are now engaging in algorithmic trading, the next step naturally is autonomous, large-scale high-frequency operation by AI intelligent agents.

This is precisely why Polymarket could produce the first killer application at the intersection of cryptocurrency and AI intelligent agents. In a market characterized by high liquidity, event-driven outcomes, and binary results, autonomous agents can operate with precision; they can absorb world events, social sentiments, and real-time inferential data, identify mispriced outcomes, and execute trades without human intervention. When this application reaches a breakthrough, it will no longer be just a crypto product. It will be the moment when autonomous agent trading becomes mainstream.