Show the numbers
What the study found
Vultax matched the same outcomes on Kalshi and Polymarket and walked both order books every five minutes for 28 hours on 12 and 13 September 2026, charging both venues' fees at the size the books held.
| Market class | Readings | Crossed before fees | Cleared fees at depth | Typical result when it did |
|---|---|---|---|---|
| NFL, before kickoff | 7,518 | 0.0% | 0.0% | — |
| College football, before kickoff | 3,220 | 3.4% | 0.1% | $0.21, once |
| Premier League, before kickoff | 1,153 | 0.3% | 0.0% | — |
| Midterms | 13,440 | 34.3% | 10.1% | $40.55 on $7,690 |
| Fed September decision | 1,011 | 38.5% | 33.3% | $714.70 on $325,669 |
In-play sports readings are excluded: the two venues' books were read up to two minutes apart. Results exclude the cost of capital, transfers and settlement risk.
How the arithmetic works
Buying Yes on one venue and No on the other for the same outcome pays exactly $1 per pair whichever way the event goes, as long as both contracts settle on the same facts. The trade is worth doing only if the two prices add up to less than $1 after everything it costs.
- Polymarket charges takers rate × C × p × (1 − p), with the rate set by market type; makers pay nothing. Rates as published in Polymarket's fee documentation on 12 September 2026; the U.S. exchange uses 0.06.
- Kalshi's API lists most series under a quadratic fee with a multiplier; the 0.07 base rate is the figure reported by independent fee trackers, because Kalshi's own schedule could not be retrieved from our environment. Kalshi rounds trade fees up, which weighs on small orders.
- Money in both legs is tied up until the later settlement. The calculator charges it at the rate you enter, pro rata by day.
- Not included: price moving while you place the second leg, partial fills, and contracts that turn out to resolve on different rules. Those are the costs that usually decide the trade.
This page illustrates arithmetic. It is not a recommendation to trade and it is not financial advice.