2025 Prediction Market Review: Total Trading Volume Exceeds $50 Billion, Top Two Players Hold Over 97.5% Market Share

Odaily星球日报Опубликовано 2026-01-08Обновлено 2026-01-08

Введение

2025 Prediction Market Review: Total Trading Volume Exceeds $50 Billion, Duopoly Holds Over 97.5% Market Share In 2025, the prediction market sector reached a total trading volume of $50.25 billion, with Kalshi and Polymarket dominating the landscape by capturing over 97.5% of the market share. The remaining ecosystem accounted for just $1.25 billion in trading volume, involving emerging platforms like Azuro, TrendleFi, and others. Kalshi reported a record-breaking year with a nominal trading volume of $23.8 billion, marking a 1,108% year-over-year increase. December alone saw $6.38 billion in trades, setting multiple historical highs. Polymarket’s annual volume was estimated at approximately $22 billion, though data varied across sources like DefiLlama and Dune. Sports dominated Kalshi’s trading volume (85%), while Polymarket saw a more diversified distribution: sports (39%), politics (34%), and cryptocurrency (18%). Other categories like economics and tech/science also saw significant growth, with open interest expanding substantially throughout the year. Trading activity surged in the second half of 2025, driven by major events and elections. November and December were particularly strong, with combined monthly volumes for Kalshi and Polymarket nearing $10 billion and exceeding $13 billion, respectively. The top platforms by historical trading volume were Kalshi ($27.1B), Polymarket ($23.2B), Opinion ($13.1B), Limitless ($512M), and Azuro ($444M). The sector is poised...

Original|Odaily Planet Daily(@OdailyChina)

Author|Wenser(@wenser 2010 )

As 2025 concludes, the prediction market sector has delivered its report card.

According to data from PredictionIndex.xyz, the total trading volume in the prediction market sector for 2025 reached $50.25 billion. Excluding Kalshi and Polymarket, the trading volume for the rest of the ecosystem was $1.25 billion. Additionally, data released by KalshiData shows that Kalshi's total nominal trading volume for the year reached $23.8 billion, a year-on-year increase of 1108%. With both Polymarket and Kalshi achieving valuations exceeding $10 billion, these results undoubtedly provide a strong boost of confidence to the capital markets and user base. In 2026, the prediction market sector will continue to be a focal point in crypto.

Odaily Planet Daily will provide a梳理 (overview) and brief analysis of the overall 2025 prediction market data in this article for readers' reference.

Overall Scale of the 2025 Prediction Market: The $50 Billion Level, Kalshi and Polymarket Hold Over 97.5% Market Share

PredictionIndex.xyz conducted end-to-end index tracking of the prediction market ecosystem, covering markets, infrastructure, end-users, and new experimental projects. The final results are as follows—

  • According to incomplete statistics from the platform, the total trading volume of the prediction market in 2025 was approximately $50.25 billion.
  • Excluding Kalshi and Polymarket, the trading volume for the rest of the ecosystem was $1.25 billion.
  • This long-tail market is crucial for testing and evolving new market designs, incentives, and ideas, involving projects such as azuroprotocol, TrendleFi, hyperstiti0ns, Limitless, MyriadMarkets, overtime, footballdotfun, xodotmarket, predictonfliq, DGbet_official, and BRKTgg.

If such data shows us the current "duopoly"格局 (landscape) of the prediction market, then the 2025 annual data provided by KalshiData, under the Kalshi platform, gives us a more detailed view of the火热程度 (hotness) of the current popular sectors.

Kalshi's 2025 Report Card: Trading Volume Reaches $23.8 Billion, Over 11x Year-on-Year Growth

On January 3rd, KalshiData announced that Kalshi achieved record growth across all metrics in 2025.

In terms of nominal trading volume, the annual total reached $23.8 billion, a year-on-year increase of 1108%, approximately 12.1 times.

  • December set a monthly historical high of $6.38 billion;
  • The 4th week of December set a weekly historical high of $1.7 billion;
  • December 21st set a daily historical high of $381.7 million.

In terms of the number of trades, the annual total reached 97 million, a year-on-year increase of 1680%, approximately 17.8 times.

  • December saw 27.67 million trades;
  • The 4th week of December saw 7.6 million trades;
  • December 21st saw 1.5 million trades, all setting historical highs.

In terms of open interest, the total reached $225 million, a year-on-year increase of 169%, approximately 2.7 times.

  • March 9th set a daily historical high of $533 million,
  • The 1st week of March set a weekly historical high of $530 million,
  • February set a monthly historical high of $499.5 million.

In terms of the number of contracts traded, the annual total reached 23.8 billion, a year-on-year increase of 1108%, approximately 12.1 times.

  • December set a monthly historical high with 6.38 billion contracts;
  • The 4th week of December set a weekly historical high with 1.7 billion contracts;
  • December 21st set a daily historical high with 382 million contracts.

According to data from the KalshiData website, since its launch on June 28, 2021, Kalshi's historical total trading volume has reached $2,725,575,718, with a daily average trading volume of $16,619,364, and a total number of trades of 27,242,274,566.

As for Polymarket, although it promotes the concept of an "on-chain prediction market," perhaps due to differences in statistical calibers and trading data channels, its 2025 trading volume data is relatively模糊 (vague).

Polymarket's 2025 Report Card: Trading Volume Estimated Around $22~25 Billion

According to DefiLlama data, Polymarket's annual DEX trading volume for 2025 was approximately $10.5 billion.

According to information from a Dune dashboard, Polymarket's annual trading volume for 2025 was approximately $22.5 billion.

According to information from the PredictionIndex.xyz website, Polymarket's historical cumulative trading volume is $23.2 billion.

According to the "Prediction Market Report" jointly issued by Keyrock and platforms like Dune, the total transaction volume of the prediction market in 2025 was $44 billion, of which Polymarket's trading volume was approximately $21.5 billion (Odaily Planet Daily Note: In contrast, Kalshi's trading volume was about $17.1 billion).

综合以上信息以及 Kalshi 官方平台在前文中给出的 “Kalshi 年度交易量为 238 亿美元”的数据来看,我们取一个相对中间值,预估 Polymarket 2025 年总体交易量约为 220 亿美元左右。 (Synthesizing the above information and the data "Kalshi's annual trading volume is $23.8 billion" given by the official Kalshi platform earlier, we take a relatively median value and estimate Polymarket's total trading volume for 2025 to be approximately $22 billion.)

Prediction Market Segments: Sports Rank First, Political Events and Crypto Rank Second and Third

According to relevant information from the "Prediction Market Report" jointly issued by Keyrock and platforms like Dune, specifically regarding prediction market betting events,

In 2025, Kalshi still focused on sports, with sports accounting for about 85% of the nominal trading volume.

In contrast, Polymarket exhibited a more diversified portfolio—

Sports (39%), politics (34%), and cryptocurrency (18%) together drove over 90% of prediction market betting activity.

Furthermore, calculated by trading volume,

  • Trading volume for prediction events in the economics field grew 905%, reaching $112 million;
  • Trading volume for prediction events in technology and science grew 1637%, reaching $123 million;
  • Open betting events were led by economics (grew 7x to a market size of about $800 million) and social & culture (grew 6x to a market size of about $700 million), showing increasing use for macro hedging and long-term positioning.
  • Other categories (e.g., culture, society) also saw significant growth, with the overall contract opening trading size growing from about $3.3 billion at the beginning of 2025 to around $13 billion, indicating a substantial improvement in market depth and liquidity.

Prediction Markets Also Have "Peak and Off-Peak Seasons": The Second Half of the Year Saw a Trading Volume Surge

It is worth noting that, as of late August 2025, data showed that Polymarket's trading volume was still maintained at a market size of around $7.5~8 billion. However, with the surge of various unexpected events, political events, and sporting events in the second half of the year, both Polymarket and Kalshi experienced a "trading explosion period":

September: Combined trading volume of Kalshi and Polymarket reached $1.44 billion;

October: Prediction market trading volume reached $8.7 billion, with Kalshi leading and Polymarket following;

November: Combined trading volume of Kalshi and Polymarket approached $10 billion,其中 (among which), Kalshi's trading volume reached $5.8 billion, a month-on-month increase of 32%; Polymarket's trading volume reached $3.74 billion, a month-on-month increase of 23.8%.

In December 2025, analyst Patrick Scott stated that in November 2025, prediction market trading volume exceeded $13 billion, more than three times the trading volume during the peak of the 2024 election. Polymarket, Kalshi, and OPINION accounted for the vast majority of the trading volume. Binary options are now applied to political speeches, sporting events, and listed company earnings reports, becoming a probability layer for world events and news.

Combined with Kalshi's December trading volume of $6.38 billion, Polymarket's trading volume势必同样不遑多让 (must surely be equally impressive). The overall prediction market trading volume in December 2025 might have reached a scale of $13-15 billion.

Top 5 Star Players in the Prediction Market: Kalshi, Polymarket, and Others

Finally, we can refer to the existing statistical data on the PredictionIndex.xyz website. As of the time of writing, the Top 5 platforms by prediction market trading volume are the following players:

  • Kalshi, platform historical trading volume approximately $27.1 billion;
  • Polymarket, platform historical trading volume approximately $23.2 billion;
  • Opinion, platform historical trading volume approximately $13.1 billion (Odaily Planet Daily Note: Considering the platform's launch time and platform mechanisms, there may be issues with wash trading. Here we only use the data statistics platform information as a reference);
  • Limitless, platform historical trading volume approximately $512 million;
  • Azuro, platform historical trading volume approximately $444 million.

Previously, Kalshi's CEO stated that the prediction market size is around $150 billion. Although the final total trading volume for 2025 was only one-third of what he mentioned, considering that the prediction market sector was not even fully formed in 2024, his statement of a "$150 billion market size" is not空谈 (empty talk). Combined with the certainty of major prediction events in 2026 such as the US midterm elections and the World Cup and other large sporting events, the overall market size of the prediction market in 2026 may still maintain 10x growth.

Of course, as we mentioned earlier in "Only One in Ten Prediction Markets Will Survive Until the End of the Year, Not an Exaggeration," a hot sector does not mean all projects and platforms will thrive. For ordinary players like us, considering liquidity depth and capital allocation constraints, perhaps specializing in 1-3 mainstream platforms is a better choice.

Связанные с этим вопросы

QWhat was the total trading volume of the prediction market in 2025, and what percentage of this did the top two platforms (Kalshi and Polymarket) hold?

AThe total trading volume of the prediction market in 2025 was $50.25 billion. The top two platforms, Kalshi and Polymarket, held a combined market share of over 97.5%.

QWhat was Kalshi's year-over-year growth in nominal trading volume for 2025, and what was its total volume?

AKalshi's nominal trading volume for 2025 was $23.8 billion, representing a year-over-year growth of 1108% (approximately 11 times).

QAccording to the article, what were the top three event categories by betting activity on the Polymarket platform in 2025?

AThe top three event categories by betting activity on Polymarket in 2025 were Sports (39%), Politics (34%), and Cryptocurrency (18%).

QWhich month in 2025 set a new all-time high for monthly trading volume on Kalshi, and what was the volume?

ADecember 2025 set a new all-time high for monthly trading volume on Kalshi, with a volume of $6.38 billion.

QWhat is the estimated historical total trading volume for the Polymarket platform as mentioned in the article?

AThe estimated historical total trading volume for the Polymarket platform is approximately $23.2 billion.

Похожее

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

OpenAI engineer Weng Jiayi's "Heuristic Learning" experiments propose a new paradigm for Agentic AI, suggesting that intelligent agents can improve not just by training neural networks, but also by autonomously writing and refining code based on environmental feedback. In the experiment, a coding agent (powered by Codex) was tasked with developing and maintaining a programmatic strategy for the Atari game Breakout. Starting from a basic prompt, the agent iteratively wrote code, ran the game, analyzed logs and video replays to identify failures, and then modified the code. Through this engineering loop of "code-run-debug-update," it evolved a pure Python heuristic strategy that achieved a perfect score of 864 in Breakout and performed competitively with deep reinforcement learning (RL) algorithms in MuJoCo control tasks like Ant and HalfCheetah. This approach, termed Heuristic Learning (HL), contrasts with Deep RL. In HL, experience is captured in readable, modifiable code, tests, logs, and configurations—a software system—rather than being encoded solely into opaque neural network weights. This offers potential advantages in explainability, auditability for safety-critical applications, easier integration of regression tests to combat catastrophic forgetting, and more efficient sample use in early learning stages, as demonstrated in broader tests on 57 Atari games. However, the blog acknowledges clear limitations. Programmatic strategies struggle with tasks requiring long-horizon planning or complex perception (e.g., Montezuma's Revenge), areas where neural networks excel. The future vision is a hybrid architecture: specialized neural networks for fast perception (System 1), HL systems for rules, safety, and local recovery (also System 1), and LLM agents providing high-level feedback and learning from the HL system's data (System 2). The core proposition is that in the era of capable coding agents, a significant portion of an AI's learned experience could be maintained as an auditable, evolving software system.

marsbit48 мин. назад

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

marsbit48 мин. назад

Your Claude Will Dream Tonight, Don't Disturb It

This article explores the recent phenomenon of AI companies increasingly using anthropomorphic language—like "thinking," "memory," "hallucination," and now "dreaming"—to describe machine learning processes. Focusing on Anthropic's newly announced "Dreaming" feature for its Claude Agent platform, the piece explains that this function is essentially an automated, offline batch processing of an agent's operational logs. It analyzes past task sessions to identify patterns, optimize future actions, and consolidate learnings into a persistent memory system, akin to a form of reinforcement learning and self-correction. The article draws parallels to similar features in other AI agent systems like Hermes Agent and OpenClaw, which also implement mechanisms for reviewing historical data, extracting reusable "skills," and strengthening long-term memory. It notes a key difference from human dreaming: these AI "dreams" still consume computational resources and user tokens. Further context is provided by discussing the technical challenges of managing AI "memory" or context, highlighting the computational expense of large context windows and innovations like Subquadratic's new model claiming drastically longer contexts. The core critique argues that this strategic use of human-centric vocabulary does more than market products; it subtly reshapes user perception. By framing algorithms with terms associated with consciousness, companies blur the line between tool and autonomous entity. This linguistic shift can influence user expectations, tolerance for errors, and even perceptions of responsibility when systems fail, potentially diverting scrutiny from the companies and engineers behind the technology. The article concludes by speculating that terms like "daydreaming" for predictive task simulation might be next, continuing this trend of embedding the idea of an "inner life" into computational processes.

marsbit50 мин. назад

Your Claude Will Dream Tonight, Don't Disturb It

marsbit50 мин. назад

Торговля

Спот
Фьючерсы
活动图片