With Constantin Bürgi and Wanying Deng
Kalshi is a prediction market where people can agree to take opposite sides of bets on events. Makers post offers and wait to be matched. Takers accept existing offers rather than seeking better prices.
Using data on 300,000 traded Kalshi contracts, we find a very strong favourite–longshot type pattern. Low-priced contracts perform badly. In particular, contracts priced below 10c win far less often than they would need to for buyers to break even, and they generate very large average losses. Higher-priced contracts do much better, and in some ranges earn small average profits in our data sample.
Kalshi’s data let you separate which side of contracts was taken by Makers and Takers. The pattern of cheap contracts performing badly shows up for both groups but it is more pronounced for Takers, which fits the idea that paying for immediacy is especially costly in the longshot end of the market.
We explain our results using a model in which people with differing beliefs select into taking different positions on the market, choosing to be either a Maker or Taker, factoring in that only a certain fraction of offers end up being matched. We find that the model explains the data on Kalshi contract prices well, provided beliefs incorporate a modest distortion such that people tend to over-estimate small probabilities.
Key research finding: Kalshi prices are highly informative, but they also display a large and systematic bias: cheap contracts are overpriced in the sense that they deliver very poor average returns, especially for Takers.
Practical advice: If you trade on Kalshi, be very wary of buying low-priced “lottery ticket” contracts and remember that taking offers is typically worse value than making them.