Categorical overround — Definition | Polymarket Trade
Categorical overround occurs when the sum of implied probabilities across all outcomes exceeds 100%. This creates an arbitrage opportunity: traders can theoretically extract profit from the pricing gap.
In depth
In prediction markets, every outcome has a price. In a categorical market—one with multiple distinct, mutually exclusive outcomes such as "Which candidate will the US elect in 2028?"—there are typically multiple YES markets, one for each candidate. Each YES market has a price reflecting the market's probability estimate for that outcome. Intuitively, if you add up all the individual probabilities, they should sum to 100% because exactly one outcome must occur. However, in practice, they often sum to more than 100%. This excess is the categorical overround. For example, if Candidate A trades at 35%, Candidate B at 38%, Candidate C at 20%, and Candidate D at 12%, the total is 105%—a 5% overround. This 5% represents a built-in profit margin for the exchange or market maker.
The concept of overround originates from traditional betting markets and gambling, where bookmakers apply a commission called "vigorish" or "juice." In traditional sports betting, if you bet $100 to win $100 on either team in a game, the bookmaker's pricing ensures they profit: your combined risk exceeds the total payout, so they pocket the spread. In prediction markets like Polymarket, the overround serves the same function. But why does it matter? First, it means that no arbitrage strategy exists to perfectly hedge all outcomes and break even—there is always a cost. Second, for traders, the presence of overround means they must choose outcomes they genuinely believe are underpriced relative to the true probability and the overround. The overround is not an accident; it is how exchanges remain solvent and incentivize liquidity providers to keep the market functioning.
On Polymarket, categorical overround becomes immediately relevant when a trader considers betting on multiple outcomes in the same categorical market. Suppose a user thinks the overround is excessive and decides to "buy all outcomes" as a hedge or speculation strategy. By buying into every outcome, they would effectively be paying 105% to receive 100%—a guaranteed loss before any trading fees. Savvy traders recognize the overround and avoid such strategies. Instead, they identify specific outcomes they believe are underpriced within the overround. For instance, if the overround is 5% but one outcome appears to be priced 2% lower than its true probability, that becomes a relative value trade. Traders also monitor overround fluctuations: if it widens to 8%, market-making activity might be declining; if it shrinks to 2%, the market has become more efficient. Polymarket users can calculate the overround themselves by summing the prices of all yes outcomes in a categorical market and checking whether the total exceeds 100%.
A common misconception is that overround is always bad for traders. In reality, overround is part of the market structure and exists for good reasons: it incentivizes market makers to provide liquidity, and it funds the exchange's operations. A trader's true objective is not to eliminate overround but to find outcomes mispriced within the overround. Another pitfall is assuming the overround is evenly distributed across all outcomes. In reality, it often concentrates on certain outcomes. For instance, longer-shot candidates might have tighter spreads and lower overround, while crowd favorites might carry excess overround due to imbalanced order flow. Traders who blindly bet on popular outcomes absorb more of the overround cost. Additionally, newcomers sometimes confuse categorical overround with the bid-ask spread. The overround is about the sum of all outcomes; the spread is the difference between buy and sell prices for a single outcome. Both exist and both cost traders, but they are separate concepts. Finally, some traders assume that if overround is low, the market is automatically "fair." This is false—a low overround can coincide with any distribution of true probabilities, and the market can still contain numerous mispricings waiting to be exploited.
Understanding categorical overround requires familiarity with several related concepts. The "implied probability" of an outcome is derived from its price: for a YES market priced at $0.35, the implied probability is 35%. The "vigorish" or "juice" is another name for the overround, expressed as a cost to traders. The "bid-ask spread" is the gap between the highest buy order and lowest sell order for a single outcome. "Arbitrage" opportunities arise when mispricings exist—if the implied probability of all outcomes sums to 105%, arbitrage traders could theoretically profit by exploiting the excess. "Market efficiency" is a related concept: highly efficient markets with many participants tend to have lower overround and fewer mispricings. Finally, the "Kelly criterion" is a position-sizing framework that many sophisticated traders use; it inherently accounts for costs like overround when determining how much to bet on a given outcome.
Example
Consider a Polymarket categorical market for 'Which party will control the U.S. Senate after the 2026 midterms?' with outcomes: Republicans 42%, Democrats 45%, and Tie 15%, totaling 102%. The 2% overround means traders collectively pay $102 to win $100, creating a cost that any strategy betting on all outcomes must absorb.