Cross-venue arbitrage checker
Same event, two venues, two prices. This works out whether the gap survives fees — each leg charged by its own venue's formula — and how much is actually locked in if you take both legs.
—
This assumes both legs fill at the shown prices and that both venues resolve the event identically. Neither is guaranteed — read the resolution criteria on both sides before committing capital, because "Fed cuts by September" can mean two different things.
Fee schedules as of 10 Jul 2026 (Polymarket), 1 Jul 2026 (Polymarket US) and Sep 2026 (Kalshi, checked 14 Sep 2026). Sources: Polymarket fees, Polymarket US fee schedule, Kalshi fee schedule. Venues revise fees; verify before trading.
How cross-venue arbitrage works
A YES contract and a NO contract on the same event are complementary: exactly one of them pays $1. If you can buy both for less than $1.00 combined, the difference is locked profit regardless of the outcome.
In a single book that never happens — YES and NO sum to $1 by construction. Across two venues it happens regularly, because the books are moved by different crowds at different speeds. That is the entire opportunity.
Why most visible gaps aren't tradeable
- Fees on both legs. Near 50¢ Kalshi's taker fee is 1.75¢ per contract and Polymarket's is 1.00–1.75¢ depending on category (crypto is the most expensive; geopolitics is free). A 2¢ gap is gone before you finish typing.
- Execution risk. Two orders on two venues aren't simultaneous. If the first fills and the second moves, you're holding a directional position you never wanted.
- Resolution mismatch. Similar titles, different criteria. This is the one that turns "risk-free" into a total loss.
- Capital lockup. Money is tied up until settlement — sometimes months. A 3% locked return over four months is not a 3% return.
The realistic version
Durable gaps of 5¢+ are rare and disappear in seconds because automated systems are watching both books continuously. Which is the honest framing of this tool: it tells you whether a gap you've spotted is worth taking, but by the time a human spots one, it's usually gone. Machines close these — that's an entire bot category.
FluxrBot isn't an arbitrage engine. It trades the gap between news and price rather than between two books — a slower-closing, larger inefficiency. See how the two venues split for why watching both matters either way.
Common questions
How does arbitrage work between Polymarket and Kalshi?
If the same event is priced differently on two venues, buying YES on the cheaper one and NO on the other locks in a profit when the combined cost is under $1.00. The gap must exceed both venues' fees to be real.
Why do prices differ between prediction markets?
Different user bases and liquidity. Kalshi's flow is sports-heavy and US-regulated; Polymarket's is crypto-native and politics-heavy. News reaches the two books at different speeds, which is exactly when gaps open.
Is prediction market arbitrage risk-free?
No. The main risks are execution (one leg fills, the other moves), differing resolution criteria between venues for events that sound identical, capital lockup until settlement, and fees eating a thin spread.
How large does the gap need to be?
Both venues charge takers a fee shaped like rate × price × (1 − price) per contract, peaking near 50¢: 1.75¢ on Kalshi, 1.00–1.75¢ on Polymarket depending on category (0 on geopolitics), 1.50¢ on Polymarket US. With a Kalshi leg and a Polymarket politics leg near 50¢ that is about 2.75¢ per pair, so a gap under roughly 3¢ is usually not tradeable. This calculator applies each venue's own formula to its own leg so you can see the real number.