Despite explosive growth and notable surges in activity surrounding major events like the 2024 election, a significant majority of individual markets on leading prediction platforms such as Polymarket and its competitor Kalshi remain remarkably shallow, characterized by low trading volumes that can pose substantial risks and raise questions about their efficacy as forecasting tools. A recent CNBC analysis has revealed that approximately 70% of all closed markets on Polymarket, since 2021 through the end of May this year, never surpassed $10,000 in reported notional volume. This prevalence of "thin markets" highlights a critical aspect of the burgeoning prediction market industry, one that has profound implications for traders, market dynamics, and the broader utility of these platforms.
Understanding the Prediction Market Landscape
Prediction markets are platforms where users bet on the outcome of future events by trading shares that represent different possibilities. For instance, in a market asking "Will Candidate A win the election?", shares might be traded for "Yes" and "No." The price of these shares, typically ranging from $0 to $1, is often interpreted as the market’s real-time probability of that event occurring. These platforms aim to harness the "wisdom of the crowd" to generate more accurate forecasts than traditional polling or expert opinions, finding applications in diverse fields from politics and finance to technology and sports.
Polymarket, launched in 2020, quickly distinguished itself with its user-friendly interface and focus on politically and culturally relevant events. Its growth trajectory has been steep, marked by periods of immense trading activity, particularly during high-stakes political cycles and significant global events. Kalshi, another prominent player, operates under a different regulatory framework, having received approval from the U.S. Commodity Futures Trading Commission (CFTC) to list event contracts. Both platforms have attracted a diverse user base, from casual observers seeking entertainment to seasoned traders looking for unique alpha opportunities or hedging strategies. The appeal lies in their dynamic nature, offering real-time insights into collective sentiment and potential future outcomes.
However, the rapid expansion of these platforms has also brought increased scrutiny, not only from financial regulators, such as the CFTC, which settled with Polymarket over unregistered offerings in 2022, but also from academics and market analysts keen to understand their internal mechanics and reliability. The volume of trading within these markets is a crucial metric, as it directly impacts liquidity, price stability, and ultimately, the confidence users can place in the market’s "predicted" probabilities.
The CNBC Analysis: A Deep Dive into Shallow Markets
The CNBC investigation, which meticulously reviewed data from Polymarket’s Gamma API spanning from 2021 to May 2026, painted a clear picture of an ecosystem dominated by low-volume activity. The Gamma API records notional volume on both sides of a trade, providing a comprehensive view of market engagement. The finding that approximately 70% of all closed markets registered under $10,000 in reported volume is striking. This indicates that while headline-grabbing events might attract millions in bets, the vast majority of markets are niche, ephemeral, or simply fail to capture significant participant interest.
Further underscoring this trend, the analysis revealed that fewer than 10% of all closed markets managed to attract between $100,000 and $1 million in reported volume. The truly high-volume markets, those exceeding $1 million, constituted an even smaller fraction. Alarmingly, over 45,000 markets, representing nearly 5% of all closed markets, had no reported volume whatsoever, suggesting either failed launches, zero interest, or highly illiquid conditions from inception.
This phenomenon is not unique to Polymarket. A separate analysis conducted on the on-chain platform Dune indicated a similar prevalence of shallow markets on Kalshi. It is important to note, however, that Kalshi’s notional volume on Dune counts only one side of the trade, making direct comparisons of absolute volume figures challenging, but the qualitative trend of numerous thinly traded markets appears consistent across both platforms. The existence of thousands of markets with minimal or no trading activity raises fundamental questions about the operational efficiency and overall market health of these platforms.
The Perils of Illiquidity: Expert Perspectives
The abundance of low-volume markets is far from ideal for prediction market traders, presenting a unique set of challenges distinct from those encountered in highly liquid financial instruments. Constantin Bürgi, a professor of economics at University College Dublin, highlighted the inherent volatility of such environments. "Thin markets by nature imply that small investments can result in large market movements and are typically more volatile," Bürgi explained to CNBC. This means that a relatively modest trade can disproportionately swing the market price, making it difficult for traders to enter or exit positions at predictable values and increasing the risk of significant, rapid losses.
Eric Zitzewitz, a professor of economics at Dartmouth College, further elaborated on the vulnerability faced by new traders in these illiquid conditions. He pointed out that "spreads between buying and selling can blow out," effectively making trades more expensive. In a liquid market, the difference between the highest bid and lowest ask (the spread) is typically narrow. In a thin market, however, this gap can widen considerably, meaning a trader might have to sell their shares at a much lower price than they could buy them, or vice versa, eroding potential profits or exacerbating losses. This increased transaction cost acts as a disincentive, particularly for retail traders with smaller capital.
Experienced traders, often prioritizing capital efficiency and predictable execution, also find thin markets less appealing. Logan Sudeith, a 26-year-old former financial risk analyst who transitioned to full-time prediction market trading last fall, articulated this preference. "I like higher volume, short term [markets]," Sudeith stated, emphasizing that such markets allow for more efficient deployment and rotation of capital. His preference for markets lasting up to a week aligns with the CNBC analysis, which found that these short-term markets had the highest number of contracts with at least $1 million in reported volume on Polymarket. These markets frequently revolved around high-interest topics such as geopolitical conflicts (e.g., the war in Iran), or figures like former U.S. President Donald Trump and Elon Musk, suggesting that immediate relevance and rapid resolution are key drivers of liquidity.
Zitzewitz echoed this sentiment, noting that participants are "more likely to trade in a market [when] lots of people are there." This reflects a fundamental aspect of market psychology: liquidity begets liquidity. Traders are drawn to markets where they can easily enter and exit positions without significantly impacting prices, creating a virtuous cycle that amplifies volume and stability. Conversely, illiquidity can create a vicious cycle, deterring potential participants and ensuring markets remain shallow.
The Bot Factor: Dominance in Thin Markets
A significant contributor to the volume, particularly in shallow markets, comes from automated trading programs, or "bots." Joshua Della Vedova, a business professor at the University of San Diego, revealed that over 80% of volume in Polymarket’s markets under $10,000 originates from bots. Della Vedova identified these digital accounts by their trading patterns, specifically wallets making more than 50 trades per day or over 1,000 total trades.
His research, based on Polymarket’s on-chain data, provided fascinating insights into bot behavior. From November 2022 to February 2026, bots reportedly generated roughly $1.2 million in shallow markets, while accumulating a far more substantial $35.1 million in markets that boasted more than $10 million in volume. This stark contrast highlights that while bots are active across the entire spectrum of markets, their ultimate goal is profit maximization, which naturally draws them to environments with greater capital and higher trading frequency.
"They are making money across all markets," Della Vedova observed, differentiating this from retail traders who frequently face losses in both shallow and heavy markets. While bots contribute significantly to volume in thinly traded markets, Della Vedova suggests they do not necessarily push prices away from fair value, as the risk of high losses in such volatile environments acts as a natural constraint. Bots, he explained, prioritize making money "per transaction," and therefore, "they prefer to trade in these larger markets, but they will trade across the whole spectrum." Their presence, while providing some semblance of activity, primarily serves their own profit motives rather than necessarily enhancing the overall market quality or price discovery mechanism for other participants.
Accuracy Under Scrutiny: The Debate on Thin Market Reliability
The question of whether thin markets can still be accurate predictors of future events remains a subject of ongoing debate among experts. The conventional wisdom often suggests that higher volume correlates with greater accuracy, as more participants and more capital lead to a more robust aggregation of information.
Evercore ISI strategists, after analyzing five years of completed markets on both Polymarket and Kalshi, lent credence to this view. Their findings indicated that high-volume markets indeed possess more reliable probabilities than their thinly traded counterparts. Crucially, they noted that only 8% of markets on both platforms touched $1 million in volume. This led them to conclude that "most quoted probabilities sit in the thinly traded tail – where calibration is weakest," suggesting that the majority of price signals on these platforms may be less trustworthy than the most prominent, well-funded markets.
However, not all researchers agree on a linear relationship between market size and accuracy. Theis Ingerslev Jensen, a Yale University professor of finance, posits that accuracy is primarily driven by the quality of the traders involved, rather than merely the quantity of trading. Jensen and his colleagues at the London Business School conducted research that found skilled or informed traders were responsible for the majority of the accuracy observed on Polymarket.
"Thin markets are not automatically inaccurate, but they are less reliable," Jensen told CNBC. He emphasized that the critical determinant is "whether skilled traders still have enough incentive and ability to trade" in those markets. If informed participants can still execute trades effectively, even in low-volume environments, their collective intelligence might still guide prices towards accurate probabilities. The challenge, however, is that illiquidity inherently reduces the incentive for skilled traders due to higher costs and volatility, potentially diminishing the very mechanism that Jensen identifies as crucial for accuracy. This suggests a nuanced relationship where a baseline level of liquidity might be necessary to attract and retain the "skilled traders" who ultimately drive predictive power.
Industry Reactions and Broader Implications
The findings regarding the pervasive nature of shallow markets present a complex picture for the prediction market industry. Polymarket declined to comment on CNBC’s findings, and Kalshi did not respond to requests for comment. This lack of official response leaves the broader industry’s perspective on these findings open to interpretation, though it is not uncommon for platforms to refrain from commenting on granular internal market data.
Despite the prevalence of low-volume markets, some experts believe their existence is unlikely to fundamentally alter how prediction markets operate for the general public or Wall Street, provided users understand the inherent risks. Harry Crane, a professor of statistics at Rutgers University, acknowledges the importance of considering trading volumes. However, he argues that "the lack of liquidity, on its own, does not discredit a market’s signal or make the market economically useless." His perspective suggests that even thinly traded markets can offer valuable signals, though these signals must be interpreted with a clear understanding of their limited backing.
The broader implications extend to the perception and potential regulatory future of prediction markets. If a significant portion of these markets is illiquid and potentially less reliable, it could impact their credibility as serious forecasting tools. Regulators, who are already grappling with how to classify and oversee these platforms, might view these findings as further evidence of their speculative nature rather than their utility as legitimate financial instruments. The challenge for platforms like Polymarket and Kalshi is to demonstrate that their core value proposition – accurate forecasting – is maintained across a sufficiently robust segment of their markets, even as they host a long tail of less active ones.
As the prediction market sector continues its rapid expansion, Crane anticipates that larger, high-interest markets will likely grow further, while low-volume markets may remain shallow. This bifurcated growth trajectory underscores the need for constant vigilance from traders. "Protect yourself at all times," Crane advised, emphasizing that "each individual entity needs to address them on their own." This highlights the paramount importance of individual due diligence, risk management, and a thorough understanding of market mechanics before engaging in prediction market trading, particularly in those segments characterized by low liquidity.
Methodology: A Transparent Approach
The CNBC analysis was based on a comprehensive dataset of all closed market data obtained via Polymarket’s Gamma API, covering the period from 2021 to the end of May 2026. The Gamma API’s methodology counts notional volume on both sides of a trade, providing a holistic measure of reported market activity. To ensure the robustness and accuracy of its findings, CNBC’s analysis on Polymarket data was rigorously reviewed by Joshua Della Vedova, a respected business professor at the University of San Diego. Della Vedova independently cross-checked CNBC’s findings against his own extensive on-chain trade dataset, which comprises 222 million resolved Polymarket trades. The alignment between the two independent analyses provides a high degree of confidence in the reported statistics regarding market volumes and liquidity on Polymarket. For Kalshi, the analysis of notional volume was conducted using data from the on-chain platform Dune, with the caveat that Dune’s methodology counts only one side of the trade. This meticulous approach ensures that the insights presented are grounded in verifiable data and expert validation.
