Why is there such a large difference in the pricing of the same event between the two major prediction markets?

【Crypto World】Uniswap founder Hayden Adams recently pointed out an interesting phenomenon: the prediction market Kalshi prices the “US acquisition of Greenland” event at about 42%, while Polymarket’s pricing is only 15%-23%, and the large gap indeed warrants reflection.

Many people’s first reaction is that the difference in platform user groups leads to pricing discrepancies, but this explanation doesn’t hold water. If it were truly just differences in user structure, a trader who can access both platforms could quickly arbitrage away the price gap—that’s a reflection of market efficiency.

The actual situation is much more complex. The two platforms are not betting on the same event at all. Specifically, Polymarket’s market asks about the “probability of happening within 2026” (currently about 23%), while Kalshi asks about the “probability during Trump’s entire term” (currently about 45%). This is an essential difference in the time dimension.

In addition to the different event definitions, issues such as question phrasing, settlement condition design, oracle selection, and even the risk pricing logic of different investors can all influence the final transaction price. So what appears to be the same market actually hides multiple layers of differences—this is also the charm of prediction markets: the pricing reflects not only probabilities but also understanding of rules and risks.