A prior probability is your estimate of how likely an outcome is before you take new evidence into account.
Last reviewed 10 August 2025 · Educational, not advice
When you start to think about a question, you rarely start from nothing. A prior probability, often shortened to a prior, is your starting estimate of how likely something is based on what you already know. It might come from a base rate, such as how often this kind of event has happened in the past, or simply from informed judgement. The prior is the number you would give before fresh information arrives.
The idea comes from Bayesian reasoning, a standard way of updating beliefs. You begin with a prior, then you see new evidence, and you revise toward a new estimate called the posterior. A strong prior moves only a little when weak evidence arrives. A weak or uncertain prior moves more. The discipline is to start from a sensible base rate rather than from the latest headline, then update in proportion to how strong the new evidence really is.
Priors matter in prediction markets because a price is itself a kind of shared estimate. Reading a market well often means comparing your own prior with the implied probability in the price. If the price sits far from a solid base rate, that gap is worth understanding. Sometimes the market knows something you do not. Sometimes the crowd has overreacted to recent news and drifted away from the base rate. A clear prior helps you tell the difference.
A prior is a starting point, not a verdict. Good forecasting means holding your prior loosely enough to update when real evidence appears, but firmly enough not to be swept away by noise. None of this guarantees a result. Forecasting deals in probabilities, every outcome remains uncertain, and any position you take on the strength of a prior can still lose its full value.
Suppose an incumbent in a stable race wins about eighty percent of the time historically. That base rate is a reasonable prior, so you start near eighty percent. A credible new poll then shows a real shift. You update downward, but not all the way to the poll, because a single poll is weaker evidence than years of base rates. The prior anchored you, and the evidence moved you in proportion to its strength.
Illustrative only. Numbers are examples, not a quote or a prediction, and exclude fees.
A prior is a starting estimate, not a forecast you can rely on, and a wrong prior leads to wrong conclusions. Outcomes stay uncertain and any position can lose its full value. Stake only what you can afford to lose, never to chase a loss, and never on borrowed money. If it stops feeling like a free choice, step back. In the United States you can call or text the helpline on 1-800-GAMBLER or visit ncpgambling.org.
It is your estimate of how likely an outcome is before you consider new evidence. It often comes from a base rate or informed judgement.
You start with the prior, then update it as new evidence arrives to reach a revised estimate. Stronger evidence moves the prior more, weaker evidence moves it less.
A market price is a shared estimate. Comparing your own prior with the price helps you see whether the market may know more than you or may have overreacted to recent news.
No. A prior is only a starting point. Outcomes stay uncertain, and any position based on a prior can still resolve against you and lose its value.
The rules change fast. Get the changes that affect you, plain and current, not tips.
Independent. Every claim dated and sourced. No platform pays for its place.