Overview of Are Prediction Markets the Best Forecasting Tool Ever — or Just Another Casino?
This episode of Freakonomics Radio examines the rise of prediction markets—especially Kalshi and Polymarket—and asks whether they are genuinely better than polls, pundits, and expert opinion at forecasting the future, or whether they are simply a new form of gambling. The first half focuses on Kalshi CEO Tarek Mansour, the company’s regulatory-first approach, and the theory that markets can aggregate dispersed information more accurately than traditional institutions. It also introduces researcher and contract writer Nicole Kagan, who explains how prediction-market contracts are built to handle ambiguity, edge cases, and legal scrutiny.
Main Themes and Big Ideas
Prediction markets as information tools
- Prediction markets turn future events into tradable contracts with real money on the line.
- The basic argument: people are more truthful and more careful when they have skin in the game.
- These markets may be useful not just for headlines and elections, but for:
- macroeconomic indicators
- FDA drug approvals
- corporate decisions
- weather and climate
- culture and sports
The intellectual roots
- The episode traces the idea back to:
- Friedrich Hayek, who argued that useful knowledge is dispersed and centralized decision-making often misses it.
- The Iowa Electronic Markets, an early academic prediction market that beat polls in election forecasting.
- Robin Hanson, who has long argued that betting markets could become a general-purpose information institution across society.
Why Kalshi emphasizes regulation
- Kalshi says it spent years getting regulated before launching widely.
- Mansour frames this as a competitive advantage: users and institutions trust a federally regulated exchange more than offshore alternatives.
- The company presents itself as a neutral exchange, similar to the NYSE or Nasdaq, taking a fee on trades rather than betting against users.
Key People and Perspectives
Tarek Mansour, Kalshi CEO
- Positions Kalshi as a market for aggregating future expectations, not a casino.
- Argues that prediction markets are more accurate because:
- traders are incentivized to be right
- people self-select into markets they know something about
- market prices expose and correct errors in public narratives
- Says Kalshi’s mission is less about valuation or founder wealth and more about building a trusted forecasting infrastructure.
Nicole Kagan, head of research and contract writing
- Explains that Kalshi’s legal and operational complexity comes from writing contracts that clearly define:
- what event is being measured
- what source determines the outcome
- what happens in ambiguous or exceptional situations
- Notes that Kalshi has one of the richest publicly available transaction datasets in the U.S. because it is a federally regulated exchange.
- Describes how contracts must anticipate weird cases:
- delayed or missing government data
- canceled events
- unclear outcomes
- grammar or wording disputes in “what someone said” markets
Robin Hanson, economist and long-time prediction-market advocate
- Says the big promise is not just forecasting elections, but informing decisions at scale.
- Believes markets can outperform other information systems at similar cost.
- Offers a “decision market” vision for organizations, governments, nonprofits, and even personal choices.
Evidence and Examples Discussed
Kalshi’s historical and practical examples
- George Santos became a famous example of a market catching an improbable but real outcome.
- Kalshi and its rival Polymarket are presented as the leading prediction markets, with massive combined valuation and attention.
- A Federal Reserve analysis found Kalshi’s macro forecasts outperformed consensus estimates in some contexts.
- Kalshi also published analysis on the New York mayoral race, showing how market moves can reflect momentum and news flow.
Internal and corporate prediction markets
- The episode mentions early internal experiments at:
- Eli Lilly
- These markets showed promise but did not become standard practice, partly because managers often prefer ambiguity when plans fail.
Elections and polling
- Prediction markets can outperform polls, but they are not foolproof.
- The show cites cases where Kalshi and Polymarket got some elections badly wrong.
- Mansour stresses that a 5% probability still means one-in-twenty outcomes happen.
How Prediction Market Contracts Work
The contract-writing process
- First, Kalshi decides whether there is a strong economic reason for a market.
- Then it checks whether an existing regulated contract template can be used.
- If not, it writes a new one with:
- a precise underlying event
- a source for resolution
- payout rules
- edge-case procedures
Edge cases matter
- Examples of complicated scenarios:
- missing inflation data during a government shutdown
- whether a performance counts if someone is dancing but not clearly singing
- what to do when speech markets involve mispronunciations, livestream issues, or punctuation ambiguities
- Kalshi sometimes resolves uncertain markets to the last fair price rather than forcing a binary yes/no.
Critiques and Risks Raised
Is it a tool or just a casino?
- Critics worry prediction markets are just speculation with a sophisticated brand.
- The episode acknowledges:
- real money and speculation are central to the model
- sports-heavy volume can make the platform look a lot like gambling
- there are obvious risks around addiction and overtrading
Insider trading and manipulation
- Because traders may have nonpublic information, there are concerns about:
- insider trading
- manipulation
- conflict of interest
- Kalshi says regulation and KYC rules help reduce those risks, and that markets can remain accurate even when manipulators try to move prices.
Regulation and public-interest limits
- Kalshi says it will not offer markets on:
- terrorism
- assassination
- war
- unlawful activity
- anything deemed against the public interest by the CFTC
- The episode notes that the legality of prediction markets remains contested, with regulators and states likely to keep challenging them in part two.
Notable Takeaways
- Markets can be better than opinions when they force participants to reveal what they truly believe and reward accuracy.
- Domain expertise is not always enough; self-calibrated generalists can sometimes outperform experts at forecasting.
- Prediction markets are most useful when they inform decisions, not just when they predict headlines.
- Regulation is a core part of Kalshi’s identity, not an afterthought.
- Even supporters admit the industry is still developing and that trust, legality, and contract design will determine whether prediction markets become mainstream or remain niche.
What to Watch For in Part 2
- The regulatory battle over whether prediction markets are legitimate financial instruments or prohibited gambling products.
- Whether trust in prediction markets can scale beyond niche users.
- How governments, firms, and investors respond if these markets continue to outperform traditional forecasting methods.