Kalshi vs Polymarket: Which Prediction Market Is Right for You?

Most comparisons of Kalshi and Polymarket are working from a 2024 picture: Kalshi is the regulated American option, Polymarket is the offshore crypto one that Americans cannot legally touch. That framing is out of date, and it leads people to the wrong platform.

Two things changed. US access to Polymarket now runs through a separate CFTC-regulated venue. And the two exchanges have drifted so far apart in what they actually list that for most questions you will not have a choice between them at all.

This comparison uses live data from our own ingest, which polls both exchanges every five minutes.

The 2026 Access Rules

This is the part most guides still get wrong.

Polymarket now operates as two separate businesses:

polymarket.com, the international exchange, remains geoblocked for US IP addresses following the 2022 CFTC settlement. Visiting from the US returns a notice that trading is blocked and points you elsewhere.
polymarket.us is a separate, CFTC-regulated venue operated by QCX LLC. It launched in December 2025, removed its waitlist in May 2026, and requires full KYC.

So the honest answer to "can Americans use Polymarket" is now yes, but not the Polymarket you have read about. The US entity is a different exchange with its own liquidity, its own market list, and dollar settlement rather than USDC.

Kalshi's position is unchanged. It has been a CFTC-regulated Designated Contract Market since 2021, available in all 50 states, with segregated customer funds and 1099 tax reporting.

Two caveats worth knowing before you fund anything:

State law is still moving. Several states have challenged prediction markets through lawsuits and cease-and-desist orders, and Minnesota passed a ban effective August 2026 that is being contested in court. Federal legality does not settle the question in every state.

The two Polymarket entities are not interchangeable. Liquidity you see quoted for Polymarket in press coverage usually refers to the international exchange. Do not assume it transfers to the US venue.

Regulatory status changes faster than articles get updated, this one included. Verify current access before you deposit.

What Each Exchange Actually Lists

Here is where the comparison gets genuinely useful, and where our own data says something you will not find elsewhere.

A snapshot from our ingest on September 7, 2026:

| | Kalshi | Polymarket | |---|---|---| | Live markets tracked | 1,214 | 898 | | Cumulative volume | $98.5M | $907.7M | | 24 hour volume | $23.3M | $26.2M |

Polymarket's cumulative volume is roughly nine times Kalshi's, which reflects its head start and its dominance during major election cycles. Daily volumes are much closer, which tells you Kalshi has been closing the gap.

But the totals hide the interesting part. Here is how those markets break down by category:

Kalshi, by market count

| Category | Markets | 24h volume | |---|---|---| | Sports | 744 | $22.3M | | Entertainment | 126 | $30K | | Economics | 126 | $34K | | Elections | 63 | $85K | | Financials | 41 | $679K | | Politics | 41 | $4.5K | | Climate and Weather | 21 | $175K |

Polymarket, by market count

| Category | Markets | 24h volume | |---|---|---| | World | 415 | $15.1M | | Financials | 99 | $1.7M | | Climate and Weather | 98 | $646K | | Sports | 90 | $2.1M | | Elections | 76 | $1.3M | | Science and Technology | 72 | $2.1M | | Economics | 21 | $3.0M |

Kalshi has become a sportsbook. Sports account for 744 of its 1,214 live markets, or 61% of the catalogue, and carry $22.3M of its $23.3M in 24 hour volume. That is 96% of all activity on the exchange concentrated in one category. Every other category on Kalshi is, by volume, a rounding error.

Polymarket is a geopolitics and world events exchange. World markets are 415 of 898, with sports a distant fourth at 90.

That divergence is the single most important fact in this comparison, and it has a consequence nobody writes about.

They Barely Compete on the Same Questions

Because the two catalogues point in different directions, the number of questions listed on both exchanges in identical form is small.

Our matching engine compares every Kalshi contract against every Polymarket one, and direct question-for-question pairs are rare. Some of that is a limitation of automated matching, which we are actively improving, but a large part is real: the exchanges have specialised.

Two structural reasons make matching genuinely hard even when the underlying question is the same:

Kalshi reuses one title across an entire event. Every candidate contract in a race carries the identical title, with the actual outcome stored in a separate field. So "Who will win the next presidential election?" is the title for the Trump contract, the Newsom contract, and every other candidate.

The phrasings differ. Kalshi lists "Coco Gauff wins" where Polymarket lists "Will Coco Gauff win the 2026 Women's US Open?" Same question, very different text.

The practical takeaway: do not assume you can shop the same bet across both venues. Check first. On most questions you will find it on one exchange or the other, not both.

Fees, Accurately

Fee comparisons in most articles are wrong or oversimplified, including in earlier versions of this one.

Kalshi

Kalshi does not charge a flat per-contract fee. The trading fee follows a formula:

fee = 0.07 x contracts x price x (1 - price)

rounded up to the next cent. The price x (1 - price) term means fees peak on contracts trading near 50 cents and shrink toward either extreme. A contract at 50 cents carries the maximum fee load. One at 95 cents costs very little to trade.

Worked example, 100 contracts at 40 cents:

0.07 x 100 x 0.40 x 0.60 = $1.68

If that position resolves to $1.00, you staked $40 to win $60 and paid $1.68, about 2.8% of profit. Losing positions still incur the entry fee.

This matters more than it sounds. The common advice to "buy longshots for the asymmetric payout" runs into the fact that mid-priced contracts are exactly where Kalshi's fee bites hardest relative to the spread.

Polymarket

The international exchange charges no trading fee. The real costs are:

Spread, the implicit cost on every trade
Gas, small Polygon transaction fees, usually under a cent
On-ramp, typically 1% to 3% if you buy USDC with a card
Off-ramp, roughly 0.5% to 1% converting back to dollars

If you already hold USDC, zero fees is genuinely zero. If you are moving USD in and out for a single trade, budget 1% to 2% round trip. The US venue settles in dollars, which removes the conversion step but has its own fee schedule you should check directly.

Spreads

From the same snapshot, median bid/ask spread was 1 cent on Polymarket versus 3 cents on Kalshi.

One caveat on that number: our Polymarket ingest samples the highest-volume markets, so this compares Polymarket's most active contracts against Kalshi's full catalogue. It is not a like-for-like measurement, and the gap on comparable markets is likely narrower. Treat it as directional.

The broader liquidity point is solid though. Only 149 of Kalshi's 1,214 markets, about 12%, traded more than $1,000 in 24 hours. The long tail of any prediction market is thin, and a market being open is not the same as it being tradeable at a fair price.

Head to Head

| Feature | Kalshi | Polymarket | |---|---|---| | Regulation | CFTC-regulated DCM | .com unregulated; .us CFTC-regulated | | US access | All 50 states | Via polymarket.us only | | Trading fees | 0.07 x price x (1-price) per contract | None on .com | | Hidden costs | Minimal | Crypto on/off-ramp on .com | | Currency | USD | USDC on .com, USD on .us | | Catalogue focus | Sports (61% of markets) | World events (46% of markets) | | Live markets | ~1,200 | ~900 | | Cumulative volume | $98.5M | $907.7M | | Median spread | 3c | 1c (top markets) | | Mobile app | iOS and Android | iOS on .us | | Tax reporting | 1099 issued | Self-reporting on .com | | Settlement | Traditional clearing | On-chain (.com) |

Which One to Use

Choose Kalshi if you are trading sports, economic releases, or weather; you want the simplest legal path as a US resident; you want 1099 reporting; or you want a mature mobile app on both platforms.

Choose Polymarket if you are trading geopolitics, world events, or technology milestones; you are outside the US and want the deepest liquidity available; or you already hold USDC and want to avoid conversion costs.

Use both if you want coverage. Given how little the catalogues overlap, the honest reason to hold accounts on both is not arbitrage. It is that roughly half the interesting questions only exist on one of them.

About Arbitrage

Cross-exchange arbitrage is real but oversold.

The mechanic is simple. If the same event trades at 42 cents Yes on one venue and 51 cents Yes on another, you buy Yes on the cheap side and No on the expensive side for a combined 91 cents, and one side is guaranteed to pay $1.00. The 9 cent gap is your gross margin.

What the guides usually leave out:

The question has to exist on both venues. As shown above, that is the binding constraint far more often than pricing is.

Fees and spreads eat most of a small gap. A 3 cent divergence is usually not a trade once you cross both spreads and pay Kalshi's fee. Gaps need to be meaningfully larger than the round-trip cost.

Capital is committed on both sides until resolution, which can be months.

Resolution criteria can differ. Two exchanges asking a similar question may use different sources or cutoff times. That is not arbitrage, it is an unhedged bet on a technicality, and it is the way these trades most often go wrong.

We track prices across Kalshi, Polymarket, PredictIt, Metaculus and Manifold and flag divergences on our markets page. Use them as a starting point for your own diligence on whether two contracts really are the same question.

Getting Started

Confirm current access rules for your state before funding anything
Compare live odds across exchanges on our markets page
Sign up free for watchlists and price alerts
Start with one platform and an amount you are comfortable losing

If you want more grounding first, read What Are Prediction Markets?, or Understanding Prediction Market Odds for how to read prices as probabilities. For the fee mechanics in more depth, see our fees comparison.

Market data in this article is a live snapshot from our ingest on September 7, 2026. Volumes and catalogue composition shift, particularly around elections and sports seasons. Regulatory status changes too, so verify before you deposit.