Kalshi's Election Prediction Market Performance
Kalshi, a platform for prediction markets, saw significant activity leading up to the midterm elections, with traders exchanging over $2.5 billion. This was nearly double the trade volume of the 2024 general election markets.
Market Accuracy
Kalshi's markets correctly predicted the outcome of 95% of congressional and gubernatorial midterm races, according to a CBS News analysis. Experts attribute this success to the high participation rates and the availability of quality information.
How Kalshi Markets Work
Users trade on yes-or-no questions regarding event occurrences, with Kalshi profiting from transaction fees. For instance, if someone buys a "yes" on a candidate winning at a certain price, they profit if that candidate indeed wins.
Race Outcomes vs. Market Predictions
- Wins: The favored candidate won in 95% of House races, 91% of Senate races, and 97% of governor's races. On average, traders gave winners a 91% chance of success.
- Losses: The favored candidate lost in 23 races, all won by Republican candidates.
Market Calibration
To assess Kalshi's market calibration, we can compare the actual win rates with the predicted probabilities. In safe races, the favored candidate won as expected, though some had lower odds than suggested. In "likely" races, the favorite won slightly more often than forecasted.
In contrast, in races where the favorite was less certain, the higher-priced candidate didn't win as often as the market indicated. This suggests a lack of bias in favor of the favorite, which is a common phenomenon in prediction markets.
Considerations:
It's important to note that race outcomes are not entirely independent, and voter turnout can influence the outcome. As a financial professor pointed out, "If lots of people turn out [to vote] for the Democrats, then all of the Democrats are more likely to win."