Kalshi Polymarket Arbitrage: Cross-Venue Edge at WC26

A "France to win the World Cup" contract trades at 18¢ on Kalshi and the matching NO trades at 79¢ on Polymarket. Sum the two legs you need — YES at 18, NO at 81 to cover the other side — and they cost 99¢ to lock $1. That is a 1% gross arb, gone the moment you pay Polymarket's gas and Kalshi's taker fee. This is what real kalshi polymarket arbitrage looks like in June 2026: present, fleeting, and almost always thinner than it first appears.

That does not make it worthless. With WC26 kicking off June 11 and 48 teams trading across at least three venue types — a CFTC-regulated US exchange (Kalshi), an offshore on-chain market (Polymarket), and sharp sportsbooks (Pinnacle and friends) — the same binary outcome is priced by three different crowds with three different cost structures. Gaps open constantly. The skill is knowing which gaps are real money and which are a tax bill in disguise.

Why cross-venue price gaps appear at all

A binary event contract resolves to $1 if true, $0 if false. So a 22¢ price is the market saying "22% chance," before fees. If France-to-win is 22¢ on one venue and 19¢ on another, one crowd is simply more bullish — and that disagreement is the raw material of arbitrage.

Three structural forces keep the venues out of sync:

• Liquidity asymmetry. Kalshi's sports books are newer and thinner than Polymarket's headline World Cup market, which has carried eight-figure volume. Thin books move on small flow, so a single $5,000 order can knock a Kalshi outright 2–3¢ off Polymarket for minutes.
• US versus offshore demand. Kalshi serves US residents under CFTC oversight; Polymarket's deepest liquidity is offshore and crypto-native. American patriotic flow inflates the USA contract on Kalshi relative to Polymarket almost every tournament.
• Settlement and resolution timing. Sportsbooks grade instantly off the official result; Polymarket resolves through a UMA oracle dispute window that can lag hours. During that lag, a "decided" outcome can still trade at 96–98¢, not 100¢ — a real but risky gap.

The arithmetic of a two-leg arb

Say France-to-win YES is 18¢ on Kalshi and France-to-win NO is 80¢ on Polymarket. Buy both. Total outlay per matched pair is 98¢. If France wins, the YES leg pays $1; if France loses, the NO leg pays $1. Either way you collect $1 for 98¢ spent — a 2.04% return on capital locked, before fees.

The trap is leg sizing. You do not bet equal dollars; you bet to equalise the payout on both legs so you are flat regardless of result. Buying $500 of an 18¢ YES gives you a different payout than $500 of an 80¢ NO, and the mismatch reintroduces directional risk. The calculator below splits a fixed stake so both legs return the same dollar amount — change the prices to your live screen and watch the locked profit (or the loss the spread is hiding).

Notice how quickly the edge evaporates. At YES 18 / NO 80 you have a tidy lock. Nudge the NO leg to 82¢ — one bad fill, one thin book — and the pair costs 100¢: zero profit, and you have tied up capital for five weeks for nothing.

Reading the board across three venues

Before you can arb, you need the prices side by side, normalised to the same unit. Below is an illustrative WC26 outright snapshot — Kalshi and Polymarket in cents, Pinnacle's vig-removed line as the fair reference. (For why the raw sportsbook number is not yet "fair," see removing the vig.)

Two things jump out. France is 3¢ cheaper on Kalshi than Polymarket — the makings of a YES-Kalshi / NO-Polymarket arb. And the USA contract is 4¢ on Kalshi versus 2¢ on Polymarket, the textbook US-patriotic markup: a trader could sell USA exposure dear on Kalshi and buy it cheap on Polymarket, capturing the 2¢ spread if the resolution sources agree.

The frictions that kill 90% of paper arbs

A 1–2¢ gross spread is the starting point. Here is what stands between you and the profit.

Fees and gas

Kalshi charges a taker fee that scales with price and size — typically a fraction of a cent to roughly a cent per contract on outrights, but it is not zero, and it applies to both the entry and, depending on the contract, the settlement. Polymarket charges no trading fee but you pay network gas to deposit, trade, and withdraw USDC on its chain, plus the spread on converting fiat to USDC in the first place. A 1.5¢ paper edge on a $2,000 position is $30 — easily consumed by gas round-trips plus a Kalshi taker fee.

Withdrawal limits and capital lockup

Your capital is frozen until resolution — for an outright, that is potentially until the July 19 final. That is five-plus weeks of locked funds for a 2% return, an annualised rate that looks far less exciting once you account for the time. Kalshi withdrawals clear to a US bank in a day or two; Polymarket withdrawals are on-chain and fast but cost gas. Neither lets you redeploy instantly.

Resolution risk — the silent killer

This is the one that turns "risk-free" arbs into losses. Each venue resolves on its own rules and own source. If Kalshi grades "France to win" off the official FIFA result and Polymarket's UMA oracle is disputed, delayed, or — in an edge case — resolves on a different interpretation (an abandoned match, a contested final), your two legs can settle differently. You can lose on both sides. The probability is small, but it is exactly the tail the 1.5¢ edge is supposed to pay you for, and often does not.

How to actually trade this

If you still want the edge, treat it like a desk operation, not a lottery.

1. Demand a real margin of safety. Do not touch a sub-1.5¢ gross spread on outrights. After fees and gas you want at least 2–3¢ net to justify weeks of lockup. Match-level markets that resolve in 90 minutes can clear a thinner edge because your capital frees up fast.
2. Pre-fund both venues. The gap is gone by the time you bridge USDC mid-trade. Keep dry powder on Kalshi and Polymarket so you can hit both legs in the same minute.
3. Size the legs to equalise payout, not dollars. Use the calculator above every time. A mis-sized pair is a directional bet wearing an arbitrage costume.
4. Read both resolution rulebooks before you click. Confirm the venues grade off the same official source and that you understand each dispute window. If the rules can diverge, price that tail in or pass.
5. Prefer the gaps with a reason. The USA-on-Kalshi markup recurs every tournament. Structural, repeatable gaps beat random noise you happened to screenshot.

Before WC26's group stage thins these books further, the cleanest edges will be in match markets, not outrights: shorter lockups, faster resolution, and the same three crowds disagreeing in real time. Build the workflow now, on small size, so when a genuine 3¢ gap opens during a June 14 group game you can act in seconds instead of scrambling to fund a wallet. The traders who profit are not the ones who find the gap — everyone sees it — but the ones already positioned to capture it before it closes.