Polymarket Under Scrutiny for Potential Trading Volume Inflation

Paradigm co-founder Matt Huang has spotlighted research suggesting that the prediction market platform Polymarket could be overstating its trading volumes due to a data aggregation error. This issue may lead to double-counting across various third-party analytics tools.

Key Takeaways

Allegations have arisen that Polymarket may be inflating its trading volumes due to data aggregation errors leading to double-counting.
Research points towards a technical flaw in how trading volumes are reported across various analytics platforms.
This controversy has sparked a debate within the industry regarding measurement challenges in prediction markets.
Major data providers plan to update their dashboards to correct the reported volumes for Polymarket.
The situation highlights the competitive climate between Polymarket and its rival Kalshi as both companies work towards regulatory compliance and market growth.

Polymarket Accused of Double-Counting its Trading Volume

The investigation, led by Paradigm research partner Storm Slivkoff, indicates that public datasets and dashboards that depend on Polymarket’s reported figures may be exaggerating the platform’s actual trading activity by around 100%.

This controversy intensified when Huang shared Slivkoff’s findings on social media platform X, prompting immediate responses from Polymarket's data team. Critics suggested that Paradigm, which has invested in competitor Kalshi, was trying to undermine Polymarket through technical nuances.

Understanding the Volume Counting Discrepancy

Slivkoff’s analysis revealed that Polymarket’s smart contracts generate separate OrderFilled events for the maker and taker sides of each trade. This leads to repetitive representations of the same transactions.

Most analytics platforms determine trading volume by totaling these events, which results in counting the same transaction twice.

For example, if YES tokens are sold for $4.13, it generates two OrderFilled events for that amount, causing dashboards to inaccurately report $8.26 in volume instead of the actual $4.13 traded.

This confusion is rooted in Polymarket’s unique market structure, which includes several types of trades, such as typical swaps and split-merge operations, where participants trade USDC for either YES or NO positions.

While no individual event has incorrect data, aggregating all OrderFilled events without differentiating between maker and taker sides leads to systematic errors in reported volume and cash flow.

The issue spans both Polymarket’s CTF Exchange and its NegRisk exchange contracts, both of which exhibit the same event patterns.

Slivkoff’s findings, based on a transaction simulator and contract code review, point out that accurate measurement requires focusing on one-sided metrics, either taker-side or maker-side volume, instead of summing redundant event streams.

Polymarket Double-Counting Trading Volume - Polymarket USDC Volume Metrics Chart

When analyzed correctly, Polymarket’s trading volumes for October and November 2024 were approximately $1.25 billion each, about half of the $2.5 billion figures reported by most public analytics before adjustments.

Industry Reaction and Competitive Landscape

In response to these findings, Polymarket’s Primo Data quickly countered the claims, asserting that the official site displays notional taker volume accurately and without double-counting, according to industry standards similar to those used by Kalshi.

“This post focuses on dashboards used for tracking Polymarket volume, not the information available on Polymarket’s website,” Slivkoff clarified, highlighting that the issue impacts third-party analytics rather than Polymarket’s internal reporting.

Major data companies, including DefiLlama, Allium Labs, and Blockworks, have confirmed they will update their Polymarket dashboards to eliminate double-counting after reviewing Slivkoff’s findings.

Despite this, some analysts have defended existing data practices. Dragonfly's data lead, Hildobby, stated that many sophisticated dashboards have accounted for these distinctions since 2024, although he did acknowledge that the methodology lacked documentation until now.

The timing of the research has raised eyebrows, particularly since Paradigm invested in Kalshi, Polymarket's main competitor.

Will Sheehan from Parsec Finance criticized the analysis, suggesting it appeared slightly biased, while others pondered whether Paradigm’s financial interest was sufficiently disclosed to avoid bias concerns.

Storm responded by emphasizing that the work merely identifies genuine errors due to data complexity, rather than attributing blame. He noted that Polymarket itself is not liable for how third parties interpret its event data.

Beyond the initial volume debate, Nick Preszler of Melee Markets argued this controversy underscores wider measurement issues in prediction markets. He pointed out that low-priced contracts can lead to inflated notional volume figures relative to the actual capital at risk.

“When a user buys $10 of contracts at .1c each, they risk $10 but are credited for $10,000 in volume,” Preszler explained, advocating for other metrics like open interest and fee revenue for more precise industry comparisons.

This debate arises as Polymarket preps for its full relaunch in the U.S. after gaining regulatory approval from the CFTC and seeks a valuation between $12 billion and $15 billion.

At the same time, the company is facing backlash for plans to launch an internal market-making operation that could trade against its customers, echoing controversial practices already in use by Kalshi.