Sports dominate prediction markets, but don’t bet on that being the whole story

Less than two years ago, the biggest controversy surrounding prediction markets was their enabling of speculation on elections. Then in early 2025, with both court rulings and an administration change liberalizing federal regulation, prediction markets began offering sports contracts on their platforms, shifting the focus of the policy debate.

Sports now account for a significant share of activity on the two largest prediction market venues. Pew Research Center found that since July 2024, sports accounted for 80 percent of Kalshi’s trading volume and 39 percent of Polymarket’s.

At first glance, prediction market activity regarding sports can look like sports betting. Both involve money, uncertainty, and the possibility of winning or losing based on correctly anticipating a game.

That resemblance is at the center of a legal fight. States challenging Kalshi and Polymarket argue their sports contracts are a “distinction without a difference” from sports betting. But resemblance is not equivalence. To determine whether prediction market sports contracts are merely “sports betting under another name,” we need to look at how they actually work.

Consider a simple example. A sportsbook might offer a wager with a payout on whether the New York Yankees will win a game. A prediction market, by contrast, might offer a contract that pays if they win. Both involve the same game and allow participants to profit from anticipating its outcome. But the similarity ends there. As I explained in a previous post on prediction markets and gambling, sportsbooks and casinos generally set the odds and tilt them toward one side of the bet.

In a prediction market, participants trade contracts with one another, with prices determined by supply and demand. Those prices can change as new information arrives, and traders can exit their positions before the event is resolved. The market price can consequently reflect traders’ collective assessment of an outcome’s probability. Calling these contracts sports bets misses these important differences.

Even if prediction-market sports contracts were simply sports bets, that would not end the argument. Calling something “gambling” does not establish that it is harmful or should be prohibited. Voluntary transactions involving risk and uncertainty are not inherently problematic.

Similarly, it should not be a concern that sports now drive so much of the volume of prediction markets. Games happen constantly, have clearly defined outcomes, and generate continuous information as injuries, lineups, and game developments change expectations. Millions of people already follow sports and understand the basic questions being traded. High sports volume, then, is hardly surprising.

But volume is not the same as value. An NFL game can generate substantial trading activity in a few hours, while a contract on a Federal Reserve decision might trade far less frequently. Yet the latter could produce information of far greater significance to businesses, investors, and policymakers.

At the same time, as CEI Director of Finance Policy John Berlau points out, even sports contracts can have economic uses beyond speculation: businesses whose revenues depend on the performance of a team or sporting event could potentially use them to hedge those risks.

Sports can also be a useful testing ground for prediction market technology because they produce frequent, measurable outcomes and attract so many participants. These markets can demonstrate innovative ways to design contracts for events that are arguably more economically important.

And volume can matter in another way. It helps create the liquidity through which markets discover information. A paper by scholars from Yale and the London Business School finds that a small share of traders can account for much of the price discovery, while broader participation provides the liquidity that allows those prices to form.

The rapid expansion of sports contracts is now helping prediction markets operate at a much greater scale for all categories.  A recent study examining more than 300,000 Kalshi contracts illustrates the scale of the ecosystem, with participants trading across a vast number of contracts.

That scale matters because a larger market can support more participants, more trading, and ultimately a wider range of contracts. Sports may account for much of the activity, but that activity can help sustain a market that extends well beyond sports.

The rapid growth of sports contracts on prediction-market platforms is therefore less important for what the contracts themselves tell us than for what they can do for the broader market. A trader who comes for sports, for example, can also encounter contracts involving economic or political questions.

The same platform and underlying infrastructure can support markets across these categories. This is illustrated by the fact that according to Kalshi data, over a third of Kalshi users who trade on sports move on to trade on other markets, as well. Similarly, a Sacred Heart University poll found that a quarter of prediction market participants trade in non-sports prediction markets.

Meanwhile, prediction markets are attracting participation into areas beyond sports, with ordinary Americans gaining access to markets covering areas such as elections, monetary policy, inflation, economic data, weather, and corporate performance.

To the extent these markets existed at all in traditional financial venues, they were the near-exclusive domain of institutional investors and large corporations. Sports contracts are therefore helping prediction markets operate at a much greater scale, helping ordinary folks hedge economic risk and gain access to information about a broad set of topics.

Policymakers should therefore be wary of regulating an entire market by its most popular category. Sports may be dominating some prediction market venues. But these markets’ ability to distribute better information and enable hedging on a variety of topics remains the same.