How prediction-market arbitrage works

The exact same event is often listed on both Kalshi and Polymarket — "Will the Fed cut rates this meeting?", "Will Bitcoin close above $100k?", "Will it hit 90°F in NYC today?" — and the two books don't always agree on the price. When YES is cheaper on one platform than the other for an identical bet, buying the cheaper side is a mechanical edge that needs no forecast at all. That's cross-platform arbitrage — "line-shopping," the trick sports bettors have always used to beat the book, finally applied to prediction markets. This guide explains exactly how it works, walks an honest example, and is blunt about the part most write-ups skip: the gap is a real disagreement, not guaranteed free money. Everything here is paper trading. No real money, ever.

1. Why the same event can have two prices

A prediction-market price is an implied probability. If the YES contract on "Will the Fed cut rates this meeting?" trades at 62¢, the market is saying there's roughly a 62% chance it happens — pay 62¢, collect $1 if it does. (Cents here are illustrative — check the live book for real prices.)

Kalshi and Polymarket are two separate venues with separate crowds, separate liquidity and separate fee structures. A US-regulated exchange and a global crypto-settled book won't always have the same traders pushing the same number, so the same question can sit at 62¢ on one and 58¢ on the other at the same moment. Each book is its own best guess; neither is automatically "right." That persistent, real disagreement is the raw material of arbitrage.

2. The three steps: match → fair value → line-shop

Our free cross-platform arbitrage scanner does exactly three things, and they're worth understanding even if you never use the tool:

• 1 · Match. Find the identical event on both books — same outcome, same resolution criteria. This is the hard part and where most "arbitrage" goes wrong: two markets with similar titles can settle on different dates or different rules. The scanner pairs them by strict rule and shows nothing unless both legs are genuinely the same bet.
• 2 · Fair value. Blend the two books into one price — a reading that's usually better than either book alone, because it pools two independent crowds. That blended number is what the event is really "worth."
• 3 · Line-shop. Buy YES wherever it's cheapest — ideally below that blended fair value. The tool tells you which book and how much you'd save, and deep-links straight to both order books so you can check the live prices yourself.

Nobody else does this for prediction markets, and it's the homepage's "secret weapon." The homepage's live ticker even surfaces Fed price gaps as they appear, and the Opportunities scout rolls the day's widest gaps in with our other strongest readings.

3. A worked example

Say the same Fed-decision market shows YES at 62¢ on Kalshi and 58¢ on Polymarket. Blended, the event looks worth about 60¢. Line-shopping says: buy YES on Polymarket at 58¢ — you're paying 2¢ under fair value for the same payoff. If you also wanted exposure to NO, NO on Kalshi (the mirror of the 62¢ YES) is the cheaper of the two NO sides. The edge isn't "I know what the Fed will do" — it's "I bought the same thing for less." (Numbers illustrative — the live tool shows real prices, refreshed on load.)

4. Why it isn't free money (read this)

Every honest arbitrage piece has this section. A visible gap can shrink or vanish before it reaches you:

• Fees and spreads eat the gap. Each platform takes a cut and quotes a bid/ask. A 4¢ raw gap can be a fraction of that — or nothing — once you pay to get in and out on both sides.
• Resolution rules can differ. "The same" market on two books can settle on different sources, criteria or dates. If the legs don't truly resolve identically, it isn't arbitrage — it's two different bets that happen to look alike. Always read both books' rules, not just the titles.
• Gaps move fast and can be thin. The cheaper side may not have enough size at that price, and the gap often closes in seconds once others see it.
• You need capital on both venues. True two-legged arbitrage means funded accounts on Kalshi and Polymarket — including the crypto rails Polymarket settles on — which is its own friction and cost.

That's why our scanner is framed as a real disagreement finder, not a risk-free-profit machine — and why, on this site, you act on it in paper money first. The Kalshi vs Polymarket explainer digs into why the two books disagree in the first place.

5. Turn the gap into a bot

You don't have to sit and refresh two order books. On this site you can wire the cross-book idea into a paper-trading bot that runs 24/7, using plain-English cross-platform signals from the signal library — for example:

• Back the blended fair value — let the bot lean on the two-book average rather than either single price.
• Only act where the books disagree — fire only when the gap between Kalshi and Polymarket is wider than a threshold you choose, so you're never trading a market both books already agree on.
• Buy the cheaper side — take YES (or NO) on whichever book is line-shopping cheaper at the moment.

Each one reads in plain English and clicks straight into a bot — no code. The bot then backtests and trades live on the public leaderboard in paper money, so you can watch the idea prove out (or not) in the open — winners and losers both, with no fabricated numbers.

6. Honest caveats

• A gap is a real disagreement between two live books — a strong signal, but not a guaranteed risk-free profit once fees, spreads, size and timing are counted.
• Only the legs that truly share a resolution are arbitrage. Check both contracts' published rules and dates before treating a gap as the same bet.
• Real two-sided arbitrage needs funded accounts on both platforms; we surface the gaps, we don't move money between exchanges for you.
• This is a sandbox. Everything here is simulated paper trading — a place to test the idea, not financial advice and not real-money order placement.

New to prediction markets? Start with how it works for the mechanics of brackets, pricing, and settlement. Trading a specific category? See the weather, crypto, sports, macro, politics, stocks, energy and tropical primers.