Prediction markets have grown explosively, but the headline volume numbers are harder to trust than they appear. According to the first report, combined global prediction market trading volume rose from roughly $9 billion in 2024 to more than $40 billion across Kalshi and Polymarket in 2025. Yet those figures are assembled from definitions that differ by platform, creating the same kind of comparability problem that plagued DeFi analytics years ago.
Three volume counting methods produce very different totals. Notional volume counts the full face value of every contract traded, while cash traded volume counts only the actual price paid. Matched volume only includes trades that cleared against a counterparty, and some venues report unmatched order flow. Double counting both legs of a buyer-seller transaction can inflate markets overnight. None of this is unique to prediction markets; crypto exchanges faced the same wash-trading and reported-volume disputes last cycle.
The same issue appears in DeFi lending, where total value locked (TVL) is widely quoted but says nothing about utilization. A protocol with $2 billion in TVL and 20% utilization is a very different business from one with $600 million and 85% utilization. Rates quoted without utilization or token emissions are close to meaningless, and the analytics layer exists to normalize these differences. DeFi Rate, for example, normalizes figures so volumes, rates and market activity can be compared like for like across platforms.
The August volume data underscores why this matters. Kalshi and Polymarket recorded their first month-over-month trading volume decline in a year, with combined August volumes falling 14.5% to $45.33 billion, according to The Block. July had pushed combined volumes above $50 billion. Kalshi accounted for roughly 82% of August volume at $37.17 billion, while Polymarket and its U.S.-focused platform PolymarketUS together contributed $8.16 billion.
The drop may reflect a lack of major event catalysts, seasonal factors, or profit-taking after a record July. It may not be a long-term reversal, but it affects liquidity and pricing efficiency. Lower volumes can widen bid-ask spreads and make pricing less reliable, which matters because prediction market probabilities are increasingly used as sentiment inputs for crypto markets. Correctly reading those probabilities requires knowing which markets are deep and which are thin.
Regulatory uncertainty adds another layer. The CFTC claims exclusive federal jurisdiction over event contracts, while state gaming regulators argue sports-related contracts are gambling under state law. Courts have split, with the Third Circuit favoring federal preemption in KalshiEX LLC v. Flaherty in April 2026 and district courts in Nevada and Utah reaching the opposite conclusion. The CFTC proposed amendments to event contract regulations in June 2026. Sports contracts made up around 80% of Kalshi’s 2025 volume and 39% of Polymarket’s, so any restriction on sports event contracts could remove a substantial share of headline volume without changing the underlying technology.
Before quoting a figure, analysts now recommend asking whether volume is notional or cash traded, whether it counts both legs, and what share sits in sports contracts. For lending data, rates should be paired with utilization and base yield separated from token emissions. Prediction markets are entering their second data cycle, where definitions get standardized.
