Cross-Platform Arbitrage: Sportsbook vs Prediction Market Arbs

How cross-platform arbs work

An arbitrage opportunity exists when two platforms disagree on price enough that you can bet both sides and guarantee a profit regardless of outcome.

Say a sportsbook has a political candidate at +150 (implied 40%). On Kalshi, the same candidate's "No" contract trades at $0.55 (implied 55% the candidate loses). The implied probabilities sum to only 95% — that 5% gap is your arb.

You bet $100 on "Yes" at the sportsbook (+150, pays $250) and buy $154 of "No" contracts at $0.55 on Kalshi (pays $280 if No). Total outlay: $254. If Yes wins, you collect $250 + $154 in returned No capital... wait. You need to do this properly.

A realistic example

The arbitrage calculator handles the allocation math. Here's a clean setup:

• Sportsbook: Candidate Yes at +200 (implied 33.3%)
• Kalshi: Candidate No at $0.58 (implied 58%)
• Combined implied probability: 91.3%

With a $1,000 total stake, optimal allocation is roughly $630 on No at Kalshi and $370 on Yes at the sportsbook.

• If Yes wins: sportsbook pays $370 x 3.00 = $1,110. Profit: $110.
• If No wins: Kalshi pays $630 / 0.58 = $1,086. Profit: $86.

Guaranteed profit of at least $86 on $1,000. That's 8.6% risk-free.

Fees destroy thin arbs

Here's where theory meets reality. Kalshi charges roughly 7% on profits. That $86 Kalshi profit becomes $80. Your sportsbook payout is unaffected, but the arb shrinks.

More importantly, prediction market fees only hit one outcome. Run both scenarios through the fee calculator to see the true guaranteed return after fees.

A 2% theoretical arb becomes breakeven or negative after Kalshi fees. You generally need a 4%+ raw arb for it to survive fees, slippage, and the opportunity cost of locked capital.

Slippage and execution risk

Sportsbooks fill instantly. Prediction markets don't. By the time you execute leg one and move to leg two, the price may have shifted. A 5% arb can vanish in seconds on a fast-moving market.

Mitigation strategies:

• Execute the less liquid side first (usually the prediction market)
• Only arb markets with sufficient depth at your size
• Use limit orders on the prediction market side, then execute the sportsbook leg only if filled

When it's worth it

Cross-platform arbs are worth pursuing when:

• The raw edge is 5%+ before fees
• Both sides have enough liquidity for your size
• You can execute both legs within minutes
• Capital isn't locked for months waiting for settlement

Use the odds converter to get both sides into the same format, then plug them into the arbitrage calculator. If the number is green after fees, you have a real arb. If not, move on.

For background on how the two platform types differ, read sportsbooks vs prediction markets. And make sure you understand how prediction market fees affect your edge before sizing any arb.