Kalshi vs Polymarket
The differences that actually matter when you’re trying to arb the two exchanges against each other.
Both are prediction-market exchanges where contracts settle for $1 if you’re right and $0 if you’re wrong. The mechanics differ enough that price gaps open routinely — and that’s where cross-exchange arbitrage lives. Here’s the field guide.
at a glance
| Kalshi | Polymarket | |
|---|---|---|
| Regulation | CFTC-regulated, US-licensed (DCM) | Polymarket US: CFTC-regulated via QCEX (live for US iOS since May 2026). Original offshore: KYC-light, US-blocked |
| Settlement currency | USD (bank deposit) | USDC on Polygon |
| Trading fees | ~1% of size, tier-discounted | 0% trading fees; bridge + gas costs apply |
| Withdrawals | ACH, 1–3 business days | USDC, ~minutes |
| Strong categories | US sports props, weather, macro buckets (CPI, Fed) | Politics, crypto prices, world events, pop culture |
| Weak categories | Long-form politics, world events | Sports player props, US-specific macro |
| Orderbook depth | Tightens around event week | Higher on flagship politics; thin on long tail |
| Resolution sources | Listed source per contract | UMA optimistic oracle; community-disputable |
which exchange tends to misprice what
Kalshi’s user base skews retail-US — that means strong efficiency on big political/macro contracts but slower repricing on niche international politics and crypto. Polymarket’s user base is crypto-native and global, so it’s usually ahead on crypto and global politics but lags on US-specific macro data (CPI prints, jobs reports).
The arbs you actually find tend to cluster in the overlap: high-attention US politics during quiet news weeks, NBA / NFL futures during the regular season, fed-decision contracts in the days leading up to a meeting.
structural differences that bite
- Capital lockup— Polymarket withdraws in minutes, Kalshi can take 1–3 business days. If you’re flipping arbs, the slower side is your bottleneck.
- US availability — both venues are now US-tradable. Polymarket US (CFTC-regulated via the 2025 QCEX acquisition) launched for US residents on iOS on May 12, 2026. Cross-exchange arbs are executable from the US for the first time. State-level restrictions still apply (e.g., Minnesota).
- Resolution risk — Polymarket uses UMA optimistic oracle, which has been disputed in edge cases. Kalshi uses listed sources per contract.
- Fee asymmetry— Kalshi’s ~1% fee on entry means a 2% spread isn’t actually arb. Tellmarket computes the post-fee edge so the number you see is what you actually keep.