Key takeaways
- Kalshi recorded $37.17B in August 2026 volume against $8.16B across Polymarket International and Polymarket US combined, roughly 82% of the three-venue total.
- Kalshi told a federal court that sports event contracts made up 80% to 90% of both its listed contracts and its revenue as of February 2026. Polymarket runs about 39% sports, 32% politics and 20% crypto.
- The two venues are not the same product. Kalshi is a CFTC-designated contract market with custodial, centrally cleared funds. Polymarket International is offshore, settles onchain, and is banned for US users.
- Dr. Rajiv Sethi's wash-trade research found stretches where up to 60% of Polymarket volume came from traders transacting across wallets they controlled, which means the headline comparison rests on numbers nobody audits.
- Kalshi closed a $1B round at a $22B valuation in May 2026 and was still in advanced talks for at least $750M more at roughly $40B as of mid-September. Polymarket closed at $15B in April, including $600M from the owner of the New York Stock Exchange.
- Legal exposure runs the other way from market share. Kalshi faces 19-plus federal suits, a $36B New York claim and adverse district-court rulings, while the offshore venue's problem is access rather than preemption.
What this comparison covers, and what it cannot settle
Most head-to-head coverage of Kalshi and Polymarket freezes a picture from late 2025, when Polymarket was the larger and more talked-about venue. That picture is stale. Through winter and spring 2026 Kalshi passed Polymarket and then pulled away, and the gap widened again through the World Cup. This comparison covers volume, category mix, custody, economics and legal status as of the 2 September 2026 data cutoff in the Hodl Up Prediction Markets Report 2026, with every figure carrying its vintage. It does not rank the two on forecasting accuracy, because no accepted method for that exists yet.
The volume gap is real, and the number is softer than it looks
The Block's dashboard put August 2026 combined volume at $45.33B across the three venues, down 14.5% from a revised July baseline of roughly $52.99B. Kalshi contributed $37.17B of August, down 7.3% month on month. Polymarket International and Polymarket US together recorded $8.16B, down 36.7%. August was the first clean print after the World Cup, and Kalshi held up materially better than both Polymarket venues.
Two cautions travel with those figures. The first is vintage. The Block revised its July series upward by about $2.4B after publication, and revised June by roughly $2B before that, so any comparison has to use the same snapshot on both sides. The second is composition. Dr. Rajiv Sethi and co-authors, using volume-to-capital ratios combined with counterparty network structure, identified periods in which as much as 60% of Polymarket volume was wash trading, concentrated in the era before the platform charged fees. His summary to us was short.
"Polymarket volume claims should be treated with a bit of skepticism."
Bank of America analysts put Kalshi's share of the regulated US market near 91%. That is an estimate rather than an audited figure, and it points the same direction as the volume data.
Two of our guests go deeper into these numbers. Dr. Rajiv Sethi, an economist at Columbia University, covers the wash trading study in our episode on prediction market manipulation (watch from 24:21, “The wash trading study: 60% of Polymarket volume”). Wasim Fukase, founder and CEO of Stratium, takes up where the volume comes from in our episode on Hyperliquid prediction markets (watch from 08:43, “Where the volume comes from: $400M and the World Cup”).
The two venues sell different products
A venue that is four-fifths sports operates as a betting exchange with continuous events and small repeat tickets. That is the product profile gambling regulators recognise, and it is the segment the CFTC's own proposed rule flags as hardest to defend on public-interest grounds. Polymarket spreads wider across sports, politics and crypto, per Pew's read of The Block data.
The structural differences matter more than the volume gap for anyone deciding where to trade or what to underwrite.
| Kalshi | Polymarket International | Polymarket US | |
|---|---|---|---|
| Market model | Central limit order book, centrally cleared | Off-chain matching, onchain settlement on Polygon | Regulated US platform via the QCEX licence |
| Custody | Custodial, segregated funds | Wallet signing, CTF outcome tokens | Custodial, KYC through regulated intermediaries |
| Category mix | Sports roughly 80% of cumulative volume | Sports 39%, politics 32%, crypto 20% | Tracks the international book, smaller base |
| Monetisation | Maker-taker fees, institutional data sales, block trading | Category trading fees after the 2026 reform, data distribution via ICE | Regulated US fee schedule |
| Token | None. Equity only | No platform token launched as of 24 June 2026 | None |
Custody is the difference most retail comparisons skip. On Kalshi, funds sit with a regulated intermediary and clearing is central, so the failure mode is counterparty and regulatory. On Polymarket International, self-custody removes some exchange-failure risk while the oracle becomes existential, because a market that settles on the wrong answer is wrong no matter how deep it was.
Legal status is where the ranking inverts
Kalshi's federal licence is its main asset and its main exposure. A Third Circuit panel sided with it against New Jersey in April 2026, the first federal appellate ruling on the question. District judges in New York and Connecticut then went the other way, holding that the Commodity Exchange Act does not preempt state gambling law. New York sued for more than $36B on 31 July, and on 11 August the CFTC invoked emergency authority to order Kalshi to keep operating there. A circuit split is forming and Supreme Court review is live.
Polymarket International has a different problem. It has been banned for US users since 2022 and is blocked or restricted in more than a dozen non-US jurisdictions, most commonly under gambling law. Dr. Harry Crane of Rutgers estimated that US traders still drove about 30% of its offshore volume in the twelve months to 30 April 2026, on a range of 19% to 48%, which suggests geoblocks bind the compliant and inconvenience everyone else.
Federally, the CFTC's proposed rule on public-interest determinations remains proposed. The comment period closed on 27 July 2026, 44 state attorneys general filed against it, and the Commission has announced no timetable for finalising it.
Both venues are covered in depth in our Prediction Markets Report 2026, alongside the microstructure, integrity and regulatory analysis behind these figures. The summary of the report's main findings is the faster read.
None of this is investment advice.