Pratt's 78% odds miss the mark in LA mayoral primary

Prediction markets are now a fixed part of the midterm election cycle, but their accuracy has already drawn scrutiny. In the Los Angeles mayoral primary, traders placed millions of dollars on Spencer Pratt, a former reality television star with no prior experience in elected office.

The markets assigned Pratt a 78% probability of advancing to the November runoff. That projection proved incorrect. Pratt finished third in the primary race, a result that highlights the gap between financial speculation and actual voter choice.

Richard Witmer, a government and political science professor at Creighton University, views political trading primarily as a form of entertainment. He compares the activity to sports betting, noting that participants are essentially wagering on a specific team or candidate to secure a small financial stake in the outcome.

Witmer argues that prediction markets and traditional polls answer different questions. Political polling typically asks voters who they believe would win if the election were held today. In contrast, prediction markets ask if a specific person will win, a binary question that drives contract pricing.

Critics of the industry point to potential confusion among voters who may place too much weight on market prices. Witmer suggests that most voters rely on political party, specific issues, or economic factors like gas prices. He noted that a gallon of gas priced at four dollars likely influences voting decisions more than the predictions offered by platforms such as Kalshi or Polymarket.

The Commodity Futures Trading Commission defines a prediction market as a platform where users buy and sell contracts tied to the outcome of a future event. While these markets have grown in popularity, they face criticism regarding their fairness and accuracy.

Whether these platforms will actually shift voter behavior remains an empirical question. Witmer plans to investigate the issue post-election by asking voters directly if market data affected their decision on how to cast their ballot.