# The favorite-longshot bias: why unlikely events look likelier in markets > In betting and prediction markets, long shots are overpriced. On Kalshi, buyers of contracts under 10¢ lost over 60%. What the bias is and why it happens. Sep 28, 2026 · Markets · Sikt Intelligence · https://www.siktintelligence.com/blog/favorite-longshot-bias The **favorite-longshot bias** is one of the oldest and most reliable patterns in betting: **outcomes that are unlikely tend to be priced as more likely than they really are, and outcomes that are likely tend to be priced as less likely than they really are.** Long shots are overbet; favorites are underbet. It was first documented at the racetrack in 1949, and in 2026 it shows up clearly in modern prediction markets. ## What the bias looks like Imagine a prediction market contract that pays $1 if an event happens. If the market were perfectly [calibrated](/blog/forecast-calibration), contracts priced at 5¢ would win 5% of the time. With the favorite-longshot bias, they win *less* often than that, so the price overstates the odds. At the other end, contracts priced at 90¢ win *more* often than 90% of the time, so favorites are slightly cheap. ## The evidence: from racetracks to Kalshi **Horse racing.** Economist R. M. Griffith noted the pattern in 1949: bettors consistently overvalued long shots and undervalued favorites. It has been confirmed across decades of racetrack data since ([Snowberg and Wolfers, 2010](https://www.nber.org/system/files/working_papers/w15923/w15923.pdf)). **Prediction markets.** A 2026 study of Kalshi, the CFTC-regulated US prediction market, analysed over 300,000 contract prices from 2021 to April 2025 ([Bürgi, Deng and Whelan, 2026](https://www.karlwhelan.com/Papers/Kalshi.pdf)). It found: - Prices are **informative overall**, and get more accurate as markets approach closing. - A **clear favorite-longshot bias**: low-priced contracts win far less often than needed to break even. - **Buyers of contracts costing under 10¢ lost over 60% of their money** on average. - Contracts priced above 50¢ earned a **small positive return**. A 2026 study of 588 million trades on Polymarket also finds an overall favorite-longshot bias, with long-shot purchases under 10¢ losing substantially, though its size depends on how related contracts are grouped, and it is strongest in crypto and politics markets ([Cardozo and Rivero-Wildemauwe, 2026](https://arxiv.org/abs/2609.12878)). ## Why it happens Researchers have proposed two main explanations: 1. **Risk-love.** Some bettors enjoy the thrill of a small bet with a big payout, and accept worse odds for it. 2. **Misperceived probabilities.** People systematically overweight small chances. A 2% event *feels* more like 5% or 10%. Using a large dataset designed to separate the two, Snowberg and Wolfers found stronger support for **misperceptions**: people overestimating small probabilities, as predicted by prospect theory ([Snowberg and Wolfers, 2010](https://www.nber.org/system/files/working_papers/w15923/w15923.pdf)). The Kalshi study reached a similar conclusion: modest disagreement plus a small tendency to overstate small probabilities is enough to reproduce the pattern. ## What it means for reading market odds - **Discount long shots.** A market price of 5% on a dramatic event is often too high. - **Favorites are more reliable.** High-probability prices tend to be close, or slightly low. - **Headlines exaggerate.** Stories like "markets give X a 10% chance" usually overstate unlikely scenarios. - **Compare with evidence.** An evidence-based forecast that starts from [base rates](/blog/base-rates-forecasting) is less exposed to the thrill of the long shot. ## Why this matters for Sikt On the Sikt feed, several of the most dramatic questions show **Sikt below the market**, for example on whether two heads of state will meet, or whether it will snow in Miami. The favorite-longshot bias is one reason a careful, evidence-based forecast can sit below the market on long shots. The odds on our site are illustrative, but the principle is general: see how markets form prices in [how prediction markets work](/blog/how-prediction-markets-work), and how forecasts are judged in [the Brier score, explained](/blog/brier-score-explained). Nothing here is investment or trading advice. ## Key takeaways - The favorite-longshot bias: long shots are overpriced, favorites slightly underpriced. - It has been documented since 1949 at racetracks, is clearly present on Kalshi, and appears on Polymarket too. - On Kalshi, buyers of contracts under 10¢ lost over 60% of their money on average. - The best-supported explanation is that people overestimate small probabilities. - Treat low market prices on dramatic events with caution. ## FAQ ### What is the favorite-longshot bias? A pattern in betting and prediction markets where unlikely outcomes are priced as more likely than they are, and likely outcomes as less likely than they are. ### Does the favorite-longshot bias exist in prediction markets? Yes. A study of over 300,000 Kalshi contract prices found a clear favorite-longshot bias, and a 2026 study of Polymarket also finds one overall, though its size depends on how contracts are grouped. ### Why do people overbet long shots? Mainly because people overestimate small probabilities, and partly because some enjoy the chance of a large payout from a small stake. ## Sources - Bürgi, Deng and Whelan (2026): [Makers and Takers: The Economics of the Kalshi Prediction Market](https://www.karlwhelan.com/Papers/Kalshi.pdf), University College Dublin - Snowberg and Wolfers (2010): [Explaining the Favorite-Longshot Bias: Is it Risk-Love or Misperceptions?](https://www.nber.org/system/files/working_papers/w15923/w15923.pdf), Journal of Political Economy - Cardozo and Rivero-Wildemauwe (2026): [The Favorite-Longshot Bias in Prediction Markets: Evidence from Polymarket](https://arxiv.org/abs/2609.12878), arXiv - CEPR VoxEU: [The economics of the Kalshi prediction market](https://cepr.org/voxeu/columns/economics-kalshi-prediction-market)