Base rate — Definition | Polymarket Trade

In depth

The base rate is fundamentally simple: it's how often something actually happens. If you examine all the times a specific type of event has occurred in the past, the base rate tells you the percentage of those instances where the outcome in question came to pass. For example, the base rate of rain on a given day in Seattle might be 40%, meaning that historically, it rains about two out of every five days. The base rate strips away any special circumstances or recent signals and just answers the raw question: how frequently does this outcome actually occur?

Base rates emerge from statistics and cognitive science, and they've become increasingly important in prediction markets. The concept gained prominence partly through research on human judgment by psychologists Daniel Kahneman and Amos Tversky, who showed that people often ignore base rates when making predictions—a bias known as the base rate fallacy. In prediction markets like Polymarket, base rates matter because they represent ground truth, the actual distribution of outcomes in the real world. When a market price implies a probability that diverges sharply from the base rate, it can signal either a mispricings opportunity or important new information that the market has incorporated.

On Polymarket, a trader might encounter a market asking "Will the Fed raise rates in the next three months?" If the historical base rate of rate hikes in three-month windows is 25%, but the market price implies a 60% probability, the trader faces a decision: Is the market overpricing the event based on recent signals, or are current conditions genuinely unusual? Savvy traders use base rates as a starting point and ask themselves: What do I know about current conditions that justifies departing from the historical frequency? If they can't articulate a compelling reason for the divergence, the base rate suggests the market may be mispriced. This is especially useful on Polymarket because the platform attracts participants with varying levels of expertise, and collective mispricings relative to historical patterns can create profitable edges.

A common mistake is treating the base rate as a ceiling or floor for market prices. In reality, base rates describe the past, not the future. If underlying conditions shift or we enter a genuinely novel regime, relying too heavily on historical frequencies can be dangerous. Another pitfall is confusing the base rate with the reference class. If you're forecasting "Will this specific startup succeed?" the base rate of startup success (roughly 10%) is relevant only if your startup is truly representative of that class. A well-funded, late-stage startup operating in a proven category might have a much higher success rate than the aggregate base rate. Traders must exercise judgment about whether the base rate for a given event class actually applies to the specific question being priced.

Base rates sit within a broader ecosystem of prediction concepts. They're related to prior probabilities in Bayesian reasoning—your base rate is your prior belief before you see market prices or other new signals. They also connect to anchoring bias, the tendency to rely too heavily on the first number you see, and to the efficient market hypothesis, which posits that prices already incorporate all available information. On Polymarket, integrating base rates with other signals—recent market movement, order flow, news, and your own research—creates a fuller picture. Base rates are a tool, not a law. They become most powerful when you use them as a benchmark to measure whether the market is pricing outcomes rationally.