Prediction Markets

Betting on Elections

Prediction markets allow people to buy and sell contracts that pay out based on election outcomes. If a contract paying $1 if Candidate A wins is trading at $0.60, the market implies a 60% probability of victory. The logic is that people with real money at stake have strong incentives to incorporate all available information, making market prices accurate forecasts.

Platforms like Polymarket, Predictit (in the US), and Betfair (in the UK) have attracted significant trading volumes during elections. In 2024, Polymarket processed over $3 billion in election-related bets. Prediction market advocates argue that markets aggregate diverse information sources, including polling data, insider knowledge, and qualitative judgments, more efficiently than any single model.

Markets vs. Polls: Who Wins?

The academic evidence on whether prediction markets outperform poll-based models is mixed. Studies by Justin Wolfers and Eric Zitzewitz found that markets and poll aggregates perform similarly in most elections, with neither consistently dominating. Markets have the advantage of incorporating non-polling information and updating instantaneously. But they are susceptible to manipulation (a wealthy bettor can temporarily move prices), herd behavior (traders follow each other rather than independent analysis), and liquidity problems (thin markets can produce volatile prices).

The most notable prediction market failure was the 2016 Brexit referendum, where Betfair showed over 85% odds for Remain on the day of the vote. Markets, like polls, can be wrong when they reflect a consensus that proves unfounded.

Worked Example: Price, Probability, and Expected Value

A prediction-market price is just a probability wearing a dollar sign. A contract that pays $1 if Candidate A wins, trading at $0.60, implies the market's collective estimate that A wins is 60%. The complementary 'A loses' contract should trade near $0.40, and in an efficient market the two prices sum to about $1.00.

When do you make money? Only when your own probability estimate differs from the market's. Expected value per $0.60 contract is:

EV = (your probability of A winning) x $1.00 - $0.60

So betting is only rational when you think the crowd is wrong. This is the deep insight behind the 'wisdom of markets': to profit you must beat the consensus, and the collective effort to do so is what supposedly makes prices accurate. But the mechanism has limits. In a thin market with few traders, a single large buy of A can push the price from $0.60 to $0.70 without any new information, temporarily implying a 70% probability that reflects one deep pocket rather than genuine wisdom. That is exactly the concern raised about 2024's Polymarket, where a handful of very large accounts drove much of the volume. Read a market price as an informed estimate, then ask who is setting it and how deep the pool is.